Left Ventricular Pressure-Volume Relationships Before and After Cardiomyoplasty in Patients With Heart Failure
Background The aim of this study was to elucidate whether beneficial effects of cardiomyoplasty (CMP) in patients with dilated cardiomyopathy are the result of a decrease in existing ventricular dilatation or a prevention of further dilatation.
Methods and Results Combined micromanometer-conductance catheters were used to evaluate left ventricular pressure-volume relationships in six patients with dilated cardiomyopathy before and at 6 and 12 months after CMP. Acute changes in preload and afterload were induced by a standardized leg-tilting intervention and a bolus infusion of nitroglycerin. After CMP, end-diastolic volume (EDV) decreased from 138±10 to 103±18 mL/m2 (P<.01) at 6 months and to 83±17 mL/m2 (P<.01) at 12 months. End-diastolic pressure (EDP) decreased from 20.2±6.4 to 13.9±7.7 mm Hg (P<.01) at 6 months after CMP. Peak ejection rate and ejection fraction increased at 6 months after CMP from 594±214 to 799±214 mL/s (P<.05) and from 26.6±4.7% to 40.1±8.3% (P<.05), respectively. Peak −dP/dt decreased at 12 months after CMP from −842±142 to −712±168 mm Hg/s (P<.05). Leg-tilting before CMP increased EDP from 20.2±6.4 to 25.6±5.2 mm Hg (P<.01), end-systolic pressure (ESP) from 118±17 to 122±17 mm Hg (P<.05), and τ from 50.8±2.8 to 53.8±2.3 ms (P<.05). Six months after CMP, leg-tilting also increased EDV from 103±18 to 110±22 mL/m2 (P<.05) and ESV from 62±14 to 66±14 mL/m2 (P<.05). Before CMP, nitroglycerin decreased EDP from 20.2±6.4 to 10.4±3.8 mm Hg (P<.01), ESP from 118±17 to 96±11 mm Hg (P<.05), ESV from 100±11 to 89±7 mL/m2 (P<.05), and τ from 50.8±2.8 to 44.5±3.7 ms (P<.05). Six months after CMP, nitroglycerin decreased EDP, ESP, and τ to similar values.
Conclusions Our findings show that up to 1 year after CMP, marked decreases in left ventricular volume are present. Our measurements suggest that CMP actively reduced the dilated ventricle but did not prevent a higher EDV on an increased venous return. The latissimus dorsi muscle wrap contraction results in better synchronization of contraction and more rapid emptying of the left ventricle.
The mechanisms of possible clinically beneficial effects of CMP, as introduced by Carpentier and Chachques,1 in the treatment of heart failure have been a matter of debate for more than a decade.2 In a previous clinical study, we demonstrated a significant active contribution of the stimulated wrapped LD muscle on LV P-V relationships.3 At optimal stimulator settings, an increased contractile state of the LV LD muscle combination could be demonstrated by an increased PER, a leftward shift of the end-systolic P-V points, an increased SV, and increased peak −dP/dt. However, at clinically used cardiomyostimulator settings, no significant changes in SV or peak −dP/dt were observed. At the same time, LV systolic PER increased during LD stimulation, and an increase in EDP occurred directly after cessation of a 1:2 applied stimulation period, suggesting a decrease in workload of the heart muscle itself during LD stimulation and a dynamic effect that decreased LV EDP. Therefore, even in the nonoptimal clinical setting, CMP exerted an active hemodynamic effect on the systolic as well as the diastolic phase.
Lee et al4 and Bellotti et al5 demonstrated decreases in LV wall stress after CMP during LD stimulation in animal and patient studies, respectively. The prevention or decrease of acute or chronic ventricular dilatation may be another beneficial mechanism associated with dynamic cardiomyoplasty, as was suggested by Chachques et al6 and Kass et al.7
To evaluate these possible mechanisms, we studied LV P-V relationships in patients with dilated cardiomyopathy before CMP and at 6 and 12 months after CMP. To study the hemodynamic effects of CMP, we also performed quickly reversible changes in preload and afterload. To induce a state of acute increased preload and afterload, a standardized leg-tilting intervention was performed. An acute decrease in preload and afterload was induced by a bolus injection of NTG. The conductance catheter technique as described by Baan et al8 was used to measure instantaneous Vlv. Plv and Pao were measured simultaneously with micromanometers. The segmental volume signals of the conductance catheter were used to study wall motion.
All patients (Table 1⇓) received long-term treatment with diuretics, digoxin, and ACE inhibitors at the time of CMP. Patient 2 also received amiodarone for arrhythmias up to 6 months after CMP. The dosing schedule of these drugs were either unchanged or reduced after CMP.
The patients studied were operated in Lyon (France) and Brescia (Italy). The surgical technique has been described by Chachques et al.9 The left LD muscle was mobilized, keeping the neurovascular pedicle intact. After the attachment of two intramuscular stimulation electrodes (model SP 5528, Medtronic), the muscle was introduced into the thoracic cavity and wrapped clockwise around both ventricles. The stimulation electrodes and a sensing electrode, attached to the right ventricular wall, were connected to a cardiomyostimulator (model SP1005, Medtronic).
Preoperative and Postoperative Evaluations
The study was approved by the medical ethics committees of the two hospitals. Informed patient consent was obtained for insertion of the catheters, and for testing the cardiomyostimulator at different settings and for the use of NTG.
Patients were sedated and heparinized before catheterization. A Swan-Ganz thermodilution catheter was placed in the pulmonary artery. A dual-micromanometer transducer conductance catheter (F7, Sentron) was inserted via the femoral artery into the LV for measurement of Pao, Plv, and Vlv. The correct position of the conductance catheter was verified by fluoroscopy and inspection of the segmental conductance signals.
The conductance catheter was connected to a Leycom Sigma-5DF signal conditioner/processor (CardioDynamics, Zoetermeer, The Netherlands) to measure Vlv.8,10,11 The method is based on measurement of the time-varying electrical conductances of five segments of blood in the LV. The five segmental conductances are measured from six adjacent electrodes, and total Vlv is calculated from the sum of the five segmental conductances. The dual-field modification, which has been shown to improve the accuracy of the method, was used in all patients.12
Ventricular function analysis by Ees was performed in three patients. A change in loading conditions was induced with an 8F balloon occlusion catheter (model SP9168, Cordis), which was positioned in the inferior caval vein.
From the individual segmental conductance signals, we were able to assess wall motion abnormalities. It has been shown previously that the time-varying segmental conductances reflect time-varying segmental Vlv values as obtained by cine CT in canine hearts.13
To obtain reliable estimates of CO by thermodilution, a computer-controlled injection system was used.11 CO was determined by use of a CO computer (COM-2, Baxter) and injections of 10 mL ice-cold 5% glucose. The simultaneously recorded conductance catheter SV was calibrated by thermodilution SV.
Data Acquisition and Analysis
ECG (extremity leads), Pao, Plv, and Vlv signals were digitized at a sampling rate of 200 Hz and stored on hard disk for subsequent analysis. A dedicated data acquisition and analysis software package CONDUCT-PC (CardioDynamics) was used to analyze conductance catheter data. The following variables were calculated, among others, with the package: τ, the time constant of exponential isovolumic Plv relaxation, which was defined as the time from peak −dP/dt until the pressure reaches half its value; PER and PFR, which were calculated as maximal −dV/dt and maximal +dV/dt, respectively.
The conductance catheter measures not only the conductance of the blood inside the LV but also the conductance of the myocardium and other surrounding tissues, the parallel conductance. This parallel conductance introduces an offset term (Vc), which can be estimated by injecting 5 mL hypertonic saline (8%) into the pulmonary artery.8 This method is, in fact, an indicator dilution technique that uses the conductance of blood. The estimation of Vc was performed with the dedicated package CONDUCT-PC. The algorithm indicates the best Vc values, thus avoiding an operator-dependent bias. A reliable estimation of Vc together with matching thermodilution SV is a prerequisite for the calibration of conductance catheter volume in absolute terms.
Blood resistance, parallel conductance, and five thermodilutions were measured before all baseline measurements.
After the CMP procedures, the measurements were performed with the LD stimulator interval at the patient’s clinically used setting and amplitude in a 1:2 ratio. Episodes of 15-second duration were recorded while the patients performed expiratory breath-holding to prevent respiration-related changes in venous return.
After the baseline measurements, an increase in preload and afterload was induced by a leg-tilting procedure. The legs were passively lifted to 50 cm, and measurements during tilting were performed within 45 seconds. In three patients, an acute preload reduction was established by occluding the inferior caval vein with the balloon catheter. In all patients, the hemodynamic effects were measured 60 seconds after a 0.5 μg/kg NTG injection into the pulmonary artery.
Statistical analysis was performed with Student’s t test for paired variates, and comparison among the three data groups was also performed with randomized-block ANOVA. Values are presented as mean±SD; statistical significance was assumed at P<.05.
All data were measured during 15-second sampling periods, and the values of all beats were averaged. One patient (patient 6) died 1 week before the 12-month catheterization procedure, probably due to an arrhythmia. From another patient (patient 2) incomplete data are available due to a hemodynamic collapse during the measurement procedures at 12 months, most likely induced by the ventriculography contrast agent. This probably also affected the 12-month results of patients 1 and 3. Therefore, the results of the leg-tilting and NTG series for the 12 months after CMP are not presented.
Table 2⇓ provides baseline data for all six patients before and at 6 and 12 months after CMP. Compared with before CMP, no significant changes in HR, CI, ESP, peak +dP/dt, τ, or PFR were observed at 6 and 12 months after CMP.
EF increased from 26.6±4.7% to 40.1±8.3% (P<.05) and to 40±4% (P<.05) at 6 and 12 months after CMP, respectively. Decreases in EDV from 138±10 to 103±18 mL/m2 (P<.01) and in ESV from 101±10 to 62±14 mL/m2 (P<.01) were present 6 months after CMP. Further significant decreases compared with pre-CMP values occurred at 12 months in EDV (to 83±17 mL/m2; P<.01) and ESV (to 50±7 mL/m2; P<.01). EDP decreased from 20.2±6.4 to 13.9±7.7 mm Hg (P<.01) at 6 months after CMP, and the values at 12 months were similar to the 6-month values. Peak −dP/dt was significantly (P<.05) less negative at 12-month CMP compared with the pre-CMP values. PER increased from 594±214 to 799±214 mL/s (P<.05) at 6 months after CMP, whereas PER at 12 months after CMP was not different from the pre-CMP values. The parallel conductance component Vc was not different after CMP (299±26 versus 296±35 mL). Fig 1⇓ shows representative P-V loops of the baseline measurements of all patients before and at 6 and 12 months after CMP. An obvious leftward shift of the end-systolic P-V points after CMP is present in all patients, caused by the decreases in EDV and ESV. In Table 3⇓, results are presented of the 50-cm leg-tilting procedures before and at 6 months after CMP. In a comparison of the leg-tilting procedures before CMP with its baseline values (Table 2⇑), revealed a slightly but significantly (P<.05) increased HR; no significant change in CI, EF, EDV, ESV, peak +dP/dt, −dP/dt, PER, or PFR; but significant increases in EDP (from 20.2±6.4 to 25.6±5.2 mm Hg; P<.01), ESP (from 118±17 to 122±17 mm Hg; P<.05), and τ (from 50.8±2.8 to 53.8±2.3 ms; P<.05). A comparison of the leg-tilting procedures at 6 months after CMP with the corresponding baseline values (Table 2⇑) reveals no changes in HR, CI, EF, peak +dP/dt, peak −dP/dt, PER, and PFR but significant increases in EDV (from 103±18 to 110±22 mL/m2; P<.05), ESV (from 62±14 to 66±14 mL/m2; P<.05), ESP (from 110±12 to 119±16 mm Hg; P<.05), EDP (from 13.9±7.7 to 20.5±5.8 mm Hg; P<.01), and τ (from 49.9±2.1 to 53.9±3.2 ms; P<.05).
The results of the 0.5 μg/kg NTG injections before and 6 months after CMP are given in Table 3⇑. In a comparison of the NTG injections before CMP with baseline values, HR, CI, EF, EDV, peak +dP/dt, peak −dP/dt, PER, and PFR were unchanged, whereas significant decreases occurred in ESV (from 100±11 to 89±7 mL/m2; P<.05), EDP (from 20.2±6.4 to 10.4±3.8 mm Hg; P<.01), ESP (from 118±17 to 96±11 mm Hg; P<.05), and τ (from 50.8±2.8 to 44.5±3.7 ms; P<.05). In a comparison of the 6-month NTG data with the six corresponding baseline values (Table 2⇑), there were no changes in HR, CI, EF, EDV, ESV, peak +dP/dt, peak −dP/dt, PER, and PFR but significant decreases in EDP (from 13.9±7.7 to 8.9±4.3 mm Hg; P<.05), ESP (from 110±12 to 99±9 mm Hg; P<.05), and τ (from 49.9±2.1 to 45±3.4 ms; P<.05).
In three patients, inferior caval vein occlusion was performed before CMP as well as after CMP at the patient’s individual best LD stimulator setting. In only one patient could a sufficient decrease in loading conditions be obtained to construct an end-systolic P-V relationship. Failures were due to insufficient decreases in preload and to arrhythmias. In Fig 2⇓, the pre-CMP and 12-month post-CMP P-V relationships are presented of patient 4 during temporary inferior caval vein occlusion. Of all patients, this patient had the lowest pre-CMP EDP, which may have facilitated the decrease in EDV during the caval vein occlusion. The Ees measured before CMP was similar to the Ees of the unassisted post-CMP beats, but the end-systolic P-V relationship was shifted to the left. However, the Ees of the 1:2 LD-stimulated beats at the patient’s best stimulator setting was considerably lower than the Ees of nonstimulated beats.
In Table 4⇓, the results of the analysis of the five segmental conductance signals are presented. Preoperatively, paradoxical motions of segmental volume signals were observed in all patients in one or more segments. After the NTG bolus in five patients, the paradoxical motions changed into more synchronized segmental volume changes. After CMP, paradoxical volume changes were reduced in all patients although at 12 months after CMP, the baseline asynchrony returned in one patient. Fig 3⇓ presents an example of the segmental volume changes before and after NTG of one patient before CMP. Fig 4⇓ shows segmental volume changes before and 6 months after CMP of two patients.
This study demonstrates the potential of cardiomyoplasty to reduce long-term dilated left heart chambers in size and in EDP up to 12 months after surgery. Moreover, PER and EF increased, and the incidence of abnormal wall motion decreased. However, at 12 months after CMP, peak −dP/dt decreased significantly. On the other hand, CO, τ, and PFR were not affected.
To estimate absolute Vlv, we determined the parallel conductance factor according to the hypertonic saline method. This factor was used to calibrate intraventricular volume measured by the conductance catheter; subsequently, the conductance SV volume was matched to the mean SV of five thermodilution measurements. Next, EDV and ESV were adjusted accordingly. Possible overestimations or underestimations of thermodilution SV in the presence of severe tricuspid and/or pulmonary valve regurgitation should affect the results of this study only when a major change in these disorders occurred after CMP. Kass et al7 used contrast-ventriculography to estimate EF; with its value and while matching conductance SV to thermodilution SV, they assessed absolute LV P-V relations.
The hypertonic saline method is an indicator dilution technique and therefore independent of heart motion and shape changes due to CMP. Delahaye et al14 demonstrated in CMP patients that LD muscle wrap stimulation may have a geometric distorting influence on the heart by pulling the inferior and posterior LV wall up, which will result in inaccuracies for techniques using diameters or planes to estimate ventricular volume. The value of the parallel conductance can be obtained objectively with a dedicated computer algorithm; therefore we preferred the hypertonic saline method as an operator-independent method to measure EF to the use of echocardiography and contrast ventriculography.
In a previous study, we observed that the parallel conductance factor of LD-stimulated beats was similar to that of the unassisted beats.3 In the present study, we found comparable values before and after CMP, although the LD muscle was wrapped around the heart. This might be explained by realizing that the parallel conductance resides mainly in the ventricular wall and right ventricular blood content and is only marginally influenced by extracardiac tissues such as lungs and LD muscle.
The pronounced decreases in EDV, with a mean of 26% at 6 months and 40% at 12 months after CMP, and EDP, with a mean of 6 mm Hg, with unchanged CO, are indicative for marked changes in diastolic properties. Kass et al7 measured similar decreases in EDV and EDP, even in absolute values, in their CMP study in three patients. Jatene et al15 observed decreases in pulmonary wedge pressure in 10 patients with a mean of 8 mm Hg after CMP. With a right LD CMP in patients, Magovern et al16 observed a 18% decrease in EDV 6 months after surgery. Also, Lorusso et al17 observed a significant decrease in LV end-diastolic diameter up to 2 years after CMP. A relative dehydration status at the time of catheterization to explain the decreased EDV after CMP is not likely because most patients gained weight after CMP (Table 1⇑). Moreover, their diuretic drug regimen was essentially unaltered.
The unchanged peak −dP/dt, the unchanged time constant of Plv decline (τ), and unchanged PFR values at 6 months after CMP are in contradiction to observations by Corin et al18 in an acute CMP animal study. They observed a prolonged τ both with and without LD stimulation and a decreased peak −dP/dt due to CMP and concluded that CMP acutely hampers diastolic properties. In a previous study, we observed an increased peak −dP/dt in LD-stimulated beats at the best setting in patients evaluated up to 24 months after CMP.3 In an acute animal study with right LD CMP, Magovern et al16 did not observe any difference in τ between stimulated and nonstimulated beats. Moreover, we observed similar values of PFR before and after CMP. Therefore, CMP seems not to be associated with an impairment of diastolic properties. The finding that the peak −dP/dt at 12 months after CMP was decreased significantly compared with pre-CMP values might be related to preload dependence of this variable.19
No significant changes in peak +dP/dt, SV, and LV ESP were present after CMP, whereas PER and EF increased and ESV decreased. EF increased by 13.5% at 6 months after CMP, which is in accordance with the 12.5%, 10%, and 9% net increases as reported by Kass et al,7 Jatene et al,15 and Lorusso et al,17 respectively. Bocchi et al20 reported a significant 4% increase in EF, whereas Magovern et al16 and Delahaye et al14 reported no significant changes after right and left LD CMP, respectively.
The mean 33% increase in PER in this study can be attributed in part to systolic effects of the stimulated LD muscle, as was also demonstrated in a previous study.3 Even at nonoptimally tuned clinical settings, PER increased with a mean of 30% during assisted beats, whereas at the best stimulator settings, a mean increase of 68% was observed.
A relatively small increase in afterload and unchanged preload of the LV was the effect of the leg-tilting procedure before CMP, resulting in a sharp increase in EDP and τ. After CMP, however, EDP increased to a mean value equal to the mean baseline value before CMP. Although an increase in afterload itself may increase τ, the small rise in Pao before CMP during leg-tilting suggests that the compromised second part of the ventricular relaxation, as expressed by the increase in τ, is merely the result of an inappropriate reaction to volume loading or increased venous return.
An inability to increase SV by augmenting EDV is demonstrated in the present study through the use of the leg-tilting test before surgery. There even was a tendency toward decreased LV EDV values due to leg-tilting, albeit not significant. This abnormal preload reserve mechanism in dilated cardiomyopathy was also demonstrated by Volpe et al21 through acute isotonic volume expansion in patients in New York Heart Assocation functional classes I or II. It might be due to cross-talk between the ventricles, because dilatation of the right ventricle during rapid filling may affect filling of the LV,22 inducing a higher LV EDP at a lower LV EDV. After CMP, however, LV EDV increased during leg-tilting in all patients, which is indicative of some recruitment of the preload reserve mechanism. Comparable effects of ACE inhibitors and NTG were observed in the study of Volpe et al.21
The results of the leg-tilting procedures do not allow the conclusion that CMP prevents further acute cardiac dilatation, but the decreased EDV and EDP after CMP will at least provide a safer range. Capouya et al23 have shown in an animal study that a nonstimulated LD wrap attenuated the progression of LV enlargement. However, in the present study, an acute increase in EDV induced by an increased venous return due to leg-tilting was not prevented by the 1:2 stimulated LD muscle.
Preoperatively, the effect of leg-tilting on the diastolic phase was more pronounced than the effect on the systolic phase. After CMP, however, ESP increased significantly with a mean of 10 mm Hg due to tilting, whereas before CMP, almost no (mean, 4 mm Hg) change was observed. Together with the increased EDV, this may be indicative of improved pump function to an increased load. Thus, CMP to some extent corrected preload reserve afterload mismatch. In Fig 5⇓, the results of the leg-tilting are presented schematically with the use of mean EDP, EDV, ESV, and ESP values for the six patients before and 6 months after CMP.
In the present study, leg-tilting was considered a rather light exercise maneuver. Increases in maximum oxygen consumption during exercise testing after CMP have been observed.20 Other studies, however, failed to demonstrate such an increase.14 It can be argued that testing at the maximal exercise level is not appropriate in CMP patients because maximal exercise may evoke a high afterload and increased wall stress, τ, and EDP, overruling the possible beneficial effects of CMP.
EDP and τ decreased to similar values in both the pre- and post-CMP procedures after a standardized 0.5 μg/kg NTG injection, although the post-CMP series started at a significantly lower EDP.
NTG improved EF only before CMP, with a mean net increase of 7%; however, this change did not reach statistical significance. Sharir et al24 observed a mean 16% increase in EF due to intravenous injections of 400 μg NTG in patients with dilated cardiomyopathy; in their study, as in our series, EDV did not decrease significantly in response to NTG. Pre-CMP ESV decreased significantly, which means the increased EF is merely the result of an improvement of the systolic ejection phase. After CMP, NTG did not change EF, which points to an ejection-improving effect due to CMP that is not augmented by NTG.
ACE inhibitors, diuretics, and digoxin have become the standard therapy for dilated cardiomyopathy. ACE inhibitors attenuate progressive cardiac enlargement and improve survival.25 Beneficial effects of nitrates in patients with dilated cardiomyopathy during exercise were observed by Hecht et al.26
In patients with dilated cardiomyopathy, Hayashida et al27 demonstrated a markedly elevated wall stress throughout the cardiac cycle. Decreases in wall stress during LD stimulation were observed in animal and patient CMP studies.4,5 The net decrease in wall stress will depend on the decrease in EDV and on the contraction/relaxation cycle of the LD muscle. In the present study, Vlv values were markedly reduced after CMP, and with the addition of the 1:2 active LD muscle, the wall stress after CMP will be substantially lowered in this patient group. In an acute CMP animal study, Chen et al28 measured the pressure between LD muscle and the heart and demonstrated a considerable decrease in transmural LV pressure during LD-stimulated beats. Decreases in transmural pressure will have a wall stress–and afterload-decreasing effect during systole. The afterload-decreasing effect will be operative when the aortic valves are opened. The timing and configuration of the applied LD stimulus therefore determine the extent to which the LD muscle will exert a dynamic wall stress–and afterload-decreasing effect.
Wall Motion Asynchrony
In Fig 4⇑, the segmental conductance signals are presented of two patients before and 6 months after CMP. Before CMP, marked paradoxical movements in four ventricular volume segments can be observed, which were converted into synchronous segmental volume signals after CMP. NTG also reduced asynchronous wall motion in five of six patients during the preoperative measurement procedures (Fig 3⇑). After CMP, NTG decreased asynchrony in one of the two patients with persistent wall motion asynchrony. Asynchronous wall motion is observed even in normal ventricles and is more pronounced in patients with healed myocardial infarction.29 Nonuniformity reduces the mechanical efficiency of ventricular ejection by inducing a premature onset of relaxation and decreasing the rate of Plv fall.30 Previously, we showed that during LD stimulation at the optimal setting the rate of Plv fall increased.3 Both NTG and CMP may improve segmental synchronism by reducing myocardial wall stress, especially in the abnormal contracting segments. In the previous study, we observed in one patient that during LD muscle stimulation at the optimal setting, an aneurysmal apical segment was forced in synchrony with global volume changes only during the LD assisted beats.3
The pathophysiology of the lower Ees during optimal LD stimulation in patient 4 may reveal part of the working mechanism of CMP (Fig 2⇑). Kass et al7 observed in three patients a leftward shift of the ESPVR at 12 months after CMP, but in two patients Ees decreased and no differences were observed in Ees between assisted and unassisted beats. The leftward shift of the end-systolic P-V points in all patients after CMP in the present study also indicates a leftward shift of the ESPVR. However, in an animal study, a higher Ees was reported after CMP in an ischemic animal model, without a change between assisted and unassisted beats.31 In another animal study, increased Ees during LD-stimulated beats was reported.32
The load dependence of myocardial relaxation is a fundamental property of mammalian cardiac muscle, and evidence in the intact heart has been provided in several studies.19 Sudden increases in afterload, when applied late in the ejection phase, induce a premature pressure decline, whereas when the sudden increase in afterload is applied early in the cardiac cycle, the onset of pressure fall is delayed.33 However, unloading the ventricle during the second half of the ejection phase resulted in some extra active shortening and improved emptying by delaying the onset of relaxation.34,35
Under physiological circumstances, the ventricle shortens against a varying afterload. The stimulated LD muscle might work as a time-varying wall stress reducer and a time-varying afterload reducer. The observed worst LD stimulator settings as presented in a previous patient study occurred mostly with short delays.3 This might be due to a too-early relaxation of the LD muscle, causing a pseudoafterload increase late in the ejection phase. Also, an afterload decrease during early contraction may shorten the ejection phase.19 The best LD stimulation setting should be the moment at which maximal LD force coincides with late ejection, inducing a lowest afterload in this period and delaying relaxation. At the optimal stimulator settings, which were the settings with the longest delays, SV increased with a mean of 20% and PER increased with a mean of 68%; the subsequent initial rate of pressure fall (−dP/dtmax) was faster.3 This is analogous to the effects of an afterload decrease during the second half of the ejection phase. One might consider this effect of LD stimulation as a limiting effect on shortening deactivation of ventricular muscle. Recently, Watkins et al35 demonstrated in isolated rabbit hearts through rapid volume withdrawal at end systole the existence of shortening activation as a third mechanism that mediates cardiac contractility during ejection, together with length-dependent activation and shortening deactivation.
The lower Ees during LD stimulation at the optimal LD muscle stimulation setting of patient 4 (Fig 2⇑) can be explained by a relatively stronger afterload decrease at lower preload, because the contribution of the LD muscle may be similar at lower preload, inducing lower ESV values. Therefore, the Ees concept erroneously characterized the simultaneous actions of heart and LD muscle during LD stimulation as a lower cardiac contractile state compared with the unassisted heart beats.
In an assessment of the LV contractile state of these patients by preload-adjusted maximal power, an ejection phase index of contractile state hardly influenced by loading conditions as proposed by Kass and Beyar,36 would provide highly significant increases in contractile state after CMP because PER increased and EDV decreased significantly and LV pressure was unchanged after CMP.
Active Versus Passive Girdling Effects of CMP
During diastole, the wrapped LD muscle will exert a tension on the cardiac wall, even in total relaxed state; therefore, the measured absolute decreases in LV EDPs due to CMP implicate even lower transmural values. This passive effect may be the basis of the limiting effect of a nonstimulated LD muscle wrap to the progression of LV enlargement in an animal model.23 However, we observed in a previous study an acute significant rise in EDP when the 1:2 LD stimulation was blocked. Moreover, in some patients, a concomitant direct increase in EDV was measured, and one patient in the present study (patient 6) deteriorated each time the LD stimulator was blocked, as demonstrated by sharp increases in EDP. In all patients, PER increased during LD muscle stimulation at each stimulator setting that was applied. The increase in PER is probably explained by the active segmental synchronizing effects of the stimulated LD muscle and may contribute to the striking decreases in LV dimensions after CMP. Therefore, the so-called passive elastic girdling effect of CMP as speculated by Kass et al7 may be of secondary significance.
When making conclusions about this limited group of patients, it must be considered that patient selection, the surgical procedure, and the LD muscle conditioning protocol may vary substantially among different institutions. On the other hand, a major advantage of the present patient group was that no additional cardiac surgery was performed.
LV EDV and EDP decreased markedly after CMP, whereas CO was unchanged. Concomitant increases in PER and EF were observed. CMP was not associated with impairment of diastolic properties at 6 months after CMP because the indices of ventricular relaxation and PFR were not affected. At 12 months after CMP, however, peak −dP/dt was less negative than the pre-CMP values.
The leg-tilting maneuver resulted in unchanged EDV before CMP and in small increases after CMP, indicating some recruitment of the preload reserve mechanism and decreasing the preload reserve afterload mismatch. CMP reduced the dilated ventricle but did not prevent acute increases in LV EDV.
Nitrate administration resulted in a decrease in EDP and τ and increased wall motion synchrony.
Asynchrony in ventricular contraction of the dilated LV diminishes LV function in dilated cardiomyopathy. This asynchrony decreased in all patients after CMP, probably because the LD contraction caused more synchronization of LV contraction. The increase in synchrony might be the rationale behind the working mechanism of CMP by improving the mechanical efficiency of ventricular ejection. Because we observed immediate rises in EDP and occasionally in EDV after cessation of LD stimulation, we suppose that these results are due to an active synchronizing effect of the stimulated LD muscle. This implies that optimally tuned stimulator settings are a prerequisite for optimal segmental synchronization.
The Ees as an index for contractile state cannot be applied in cardiomyoplasty when the LD muscle is stimulated because the stimulated muscle changes the ESPVR in an unpredictable fashion.
The stimulated LD muscle might work as a time-varying wall stress and afterload reducer; the decrease in wall stress induces a more synchronized contraction/relaxation pattern, and the decrease in afterload applied at the end of the ejection phase may induce an accelerated ejection and a delayed relaxation.
Selected Abbreviations and Acronyms
|ESPVR||=||end-systolic pressure volume relationship|
|LV||=||left ventricular, ventricle|
|PER||=||peak ejection rate|
|PFR||=||peak filling rate|
|Plv||=||left ventricular pressure|
|Vc||=||parallel conductance offset term|
|Vlv||=||left ventricular volume|
|τ||=||time constant of Plv decline|
This study was supported by Telectronics Pacing Systems (Denver, Colo). We wish to thank Th. van der Nagel for excellent technical assistance and M.C. Stuart, MD, FFARCSI, for correction of the manuscript.
- Received February 23, 1997.
- Revision received June 12, 1997.
- Accepted June 19, 1997.
- Copyright © 1997 by American Heart Association
El Oakley RM, Jarvis JC. Cardiomyoplasty: a critical review of experimental and clinical results. Circulation. 1994;90:2085-2089.
Schreuder JJ, van der Veen FH, van der Velde ET, Delahaye F, Alfieri O, Jegaden O, Lorusso R, Jansen JRC, van Ommen V, Finet G, Wellens HJJ. Beat-to-beat analysis of left ventricular pressure-volume relation and stroke volume by conductance catheter and aortic model flow in cardiomyoplasty patients. Circulation. 1995;91:2010-2017.
Bellotti G, Moraes A, Bocchi E, Arie S, Medeiros C, Moreira LP, Jatene A, Pileggi F. Late effects of cardiomyoplasty on left ventricular mechanics and diastolic filling. Circulation. 1993;88:304-308.
Chachques JC, Acar C, Tapia M, Guibourt P, Fiemeyer A, Bensasson D, Berrebi A, Grare P, Bechara M, Baron JF, Carpentier M. Résultats à moyen terme de la cardiomyoplastie. Arch Mal Coeur. 1994;87:49-56.
Kass DA, Baughman KL, Pak PH, Cho PW, Levin HR, Gardner TJ, Halperin HR, Tsitlik JE, Acker AJ. Reverse remodeling from cardiomyoplasty in human heart failure: external constraint versus active assist. Circulation. 1995;91:2314-2318.
Baan J, van der Velde ET, De Bruin HG, Smeenk GJ, Koops J, Van Dijk AD, Temmerman D, Senden PJ, Buis B. Continuous measurement of left ventricular volume in animals and humans by conductance catheter. Circulation. 1984;70:812-823.
Burkhoff D, Van der Velde ET, Kass D, Baan J, Maughan WL, Sagawa K. Accuracy of volume measurement by conductance catheter in isolated, ejecting canine hearts. Circulation. 1985;72:440-447.
Steendijk P, Van der Velde ET, Baan J. Left ventricular stroke volume by single and dual excitation of the conductance catheter in dogs. Am J Physiol. 1993;264:(Heart Circ Physiol 33) H2198-H2207.
Van der Velde ET, Van Dijk AD, Steendijk P, Diethelm L, Chagas A, Lipton MJ, Glantz SA, Baan J. Left ventricular segmental volume by conductance catheter and cine-CT. Eur Heart J. 1992;13(suppl E):15-21.
Magovern JA, Park SE, Cmolik BI, Trumble DR, Christlieb IY, Magovern GJ. Early effects of right latissimus dorsi cardiomyoplasty on left ventricular function. Circulation. 1993;88:298-303.
Lorusso R, Milan E, Volterrani M, Giubbini R, van der Veen FH, Schreuder JJ, Picchioni A, Alfieri O. Cardiomyoplasty as an isolated procedure to treat refractory heart failure. Eur J Cardiothorac Surg. 1997;11:363-372.
Brutsaert DL, Sys SU. Relaxation and diastole of the heart. Physiol Rev. 1989;69:1228-1315.
Bocchi EA, Guimaraes GV, Moreira LFP, Bacal F, de Moraes AV, Barreto ACP, Wajngarten M, Belotti G, Stolf N, Jatene A, Pileggi F. Peak oxygen consumption and resting left ventricular ejection fraction changes after cardiomyoplasty at 6-month follow-up. Circulation. 1995;92(suppl II):II-216-II-222.
Volpe M, Tritto C, DeLuca N, Rubattu S, Mele AF, Lembo G, Enea I, deCampora P, Rendina V, Romano M, Trimarco B, Condorelli M. Angiotensin converting enzyme inhibition restores cardiac and hormonal responses to volume overload in patients with dilated cardiomyopathy and mild heart failure. Circulation. 1992;86:1800-1809.
Elzinga G, Van Grondelle R, Westerhof N, Van den Bos GC. Ventricular interference. Am J Physiol. 1974;226:941-947.
Sharir T, Feldman MD, Haber H, Feldman AM, Marmor A, Becker LC, Kass DA. Ventricular systolic assessment in patients with dilated cardiomyopathy by preload-adjusted maximal power. Circulation. 1994;89:2045-2053.
Hecht HS, Karahalios SE, Schnugg SJ, Ormiston JA, Hopkins JM, Rose JG, Singh BN. Improvement in supine bicycle exercise performance in refractory congestive heart failure after isosorbide dinitrate: radionuclide and hemodynamic evaluation of acute effects. Am J Cardiol. 1982;49:133-140.
Hayashida W, Kumada T, Nohara R, Tanio H, Kambayashi M, Ishikawa N, Nkamura Y, Himura Y, Kawai C. Left ventricular regional wall stress in dilated cardiomyopathy. Circulation. 1990;82:2075-2083.
Heyndrickx GR, Vantrimpont PJ, Rousseau MF, Pouleur H. Effects of asynchrony on myocardial relaxation at rest and during exercise in conscious dogs. Am J Physiol. 1988;254:H817-H822.
Nakajima H, Niinami H, Hooper TL, Hammond RL, Nakajima HO, Lu H, Ruggiero R, Thomas GA, Mocek FW, Fietsam R, Krakovsky AA, Spanta AD, Suga H, Stephenson LW, Baciewicz FA. Cardiomyoplasty: probable mechanism of effectiveness using the pressure-volume relationship. Ann Thorac Surg. 1994;57:407-415.
Sugiura S, Harada K, Yokoyama I, Momomura S, Naruse Y, Maku-Uchi H, Serizawa T, Matsunaga H, Iizuka M, Furuse A, Sugimoto T. Analysis of cardiac assistance by latissimus dorsi cardiomyoplasty with a time varying elastance model. Cardiovasc Res. 1993;27:997-1003.
Ariel Y, Gaasch WH, Bogen DK, McMahon TA. Load-dependent relaxation with systolic volume steps: servo-pump studies in the intact canine heart. Circulation. 1987;75:1287-1294.
Suga H. External mechanical work from relaxing ventricle. Am J Physiol. 1979;236:H494-H497.
Watkins MW, Higashiyama A, Chen Z, LeWinter MM. Rapid shortening during relaxation increases activation and improves systolic performance. Circulation. 1996;94:1475-1482.
Kass DA, Beyar R. Evaluation of contractile state by maximal ventricular power divided by the square of end-diastolic volume. Circulation. 1991;84:1698-1708.