|
|
||||||||
Ann Thorac Surg 2004;78:900-905
© 2004 The Society of Thoracic Surgeons
a Department of Radiology, Berlin, Germany
b Department of Cardiovascular Surgery, Charité Medical School, Humboldt-Universität, Berlin, Germany
c Department of Internal Medicine, Division of Cardiology, Emergency Hospital Berlin, Berlin, Germany
Accepted for publication December 29, 2003.
* Address reprint requests to Dr Lembcke, Institut für Radiologie, Universitätsklinikum Charité, Campus Charité Mitte, Humboldt-Universität zu Berlin, Schumannstraße 20/21, 10098 Berlin, Germany
alexander.lembcke{at}gmx.de
| Abstract |
|---|
|
|
|---|
METHODS: Ten male patients with idiopathic dilated cardiomyopathy underwent electron-beam computed tomographic (CT) examination within 1 month before, and 6 to 9 months after CSD implantation. The RV end-diastolic and end-systolic volumes (EDV, ESV) and diameters (EDD, ESD), stroke volume (SV), ejection fraction (EF), total and forward RV output (RVO, fRVO), and tricuspid regurgitation fraction (TRF) were calculated.
RESULTS: The EDV measurements decreased from 182.1 ± 49.6 to 137.5 ± 37.0 mL, ESV from 114.8 ± 47.0 to 68.3 ± 23.8 mL, EDD from 48.2 ± 6.6 to 41.6 ± 7.1 mm, and ESD from 39.6 ± 6.9 to 32.7 ± 6.5 mm (p < 0.05 for each). Ejection fraction increased from 38.5 ± 8.9 to 52.0% ± 7.7% and fRVO from 4.0 ± 0.8 to 4.6 ± 1.1 L/min (each with p < 0.05). TRF decreased from 18.2 ± 14.1 to 10.4% ± 13.5%, whereas SV and RVO remained nearly unchanged. Postoperatively, RV volumes, EF, and fRVO were not different from 15 age- and gender-matched normal control patients.
CONCLUSIONS: Implantation of a CSD leads to a decrease in RV size and improved RV performance. These data together with the results of previous studies demonstrating improved left ventricular structure and function confirm the biventricular nature of recovery with the CSD.
| Introduction |
|---|
|
|
|---|
In search of new surgical techniques for preventing heart failure progression and to reduce or delay the need for transplantation, the cardiac support device (CSD; Acorn Cardiovascular Inc., St. Paul, MN) is currently undergoing worldwide evaluation in randomized, prospective clinical trials. Implantation of this device is a modification of dynamic cardiomyoplasty and consists of positioning a compliant polyester mesh device around both ventricles instead of an actively stimulated muscle flap. The device serves to improve cardiac function by limiting the dilating capacity of the heart, reducing left ventricular wall stress and reverse remodeling of the myocardium [35]. Early results suggest that this procedure is safe to perform and that patients tend to show improvements in clinical symptoms as well as hemodynamic parameters [58]. Initial attention focused primarily on left ventricular (LV) geometry and function because they are regarded as the best predictors of long-term outcome. On the other hand, it is also known that both ventricles interact in a highly complex manner and that right ventricular (RV) function parameters likewise have a prognostic role in predicting the spontaneous course of the disease and in assessing postoperative outcome in ischemic and primary dilated cardiomyopathy [912]. Furthermore, concerns have been raised that any constrictive device could lead to more pronounced problems with RV function.
Data on RV dimensions and function are difficult to obtain due to the complicated anatomy and shape of the RV and its limited visualization by imaging modalities. However, electron-beam computed tomography (CT) is a sectional imaging modality that allows for the visualization and precise volumetric measurement of the RV with a good temporal and spatial resolution [1316].
The aim of the present study therefore was to use electron-beam CT to identify potential changes of RV geometry and function following CSD implantation.
| Material and methods |
|---|
|
|
|---|
Surgery was performed as described in detail elsewhere [57]. Briefly, after standard sternotomy and pericardiotomy, the heart size was measured and the appropriate CSD was wrapped around the ventricles and attached close to the atrioventricular groove while the heart was beating on cardiopulmonary bypass. With the ventricles fully loaded, the anterior part of the CSD was fitted, trimmed to size, and secured.
All patients of our study had congestive heart failure resulting from idiopathic dilated cardiomyopathy. Patients were in New York Heart Association (NYHA) class III or had a history of at least one NYHA class III episode, were on intensive drug therapy at the time of surgery, and had no additional systemic disease (especially no pulmonary, hepatic, or renal dysfunction). Patients with end-stage heart failure (late NYHA class IV), an unstable hemodynamic condition, or receiving intravenous inotropic agents or mechanical circulatory support were excluded from device implantation.
Because CT examination of healthy volunteers for comparison is precluded, a control group consisting of 15 retrospectively selected male patients of the same age (mean age 59.1 ± 7.4 years) was created. These controls had undergone evaluation of cardiac morphology and function for various clinical indications, but were retrospectively found to have no abnormalities. No control patient had a history of pulmonary hypertension, myocardial infarction, or congestive or valvular heart disease.
Written informed consent was obtained from all individuals after absolute or relative contraindications to electron-beam CT had been excluded. CSD implantation and CT evaluation was approved by the local ethics committee.
Electron-beam CT was performed using an established standardized protocol on an Evolution scanner (C-150 XP; GE Imatron, San Francisco, CA) in the prospectively electrocardiogram-triggered multi-slice mode (tube current 625 mA, voltage 130 kV).
After intravenous administration of contrast medium (iodine content 370 mg/mL [Ultravist 370; Schering, Berlin, Germany]) at a dose of 15 to 25 mL and a flow rate of 4 mL/s, 15 scans in two slices each at the level of the pulmonary bifurcation were acquired ("flow protocol"). Using time-density curves derived from regions of interest placed in the pulmonary artery, the time of maximal contrast medium concentration (transit time), and the forward cardiac output according to the indicator dilution method [17, 18] was calculated.
For the subsequent functional study, another 90 mL of contrast medium was injected at a flow of 3 mL/s and data were acquired during single breath-hold periods at 12 levels (slice thickness 8 mm, matrix 256x256 mm) each, with 13 scans per cardiac cycle ("cine protocol") from the base to the apex of the heart along the approximated short axis. Acquisition time per scan (temporal resolution) was 50 msec.
Postprocessing was done in the implemented evaluation mode by manually drawing the endocardial contours of the ventricles (Figs 1A and 1B) and ventricular volumes were calculated according the slice summation method.
|
In addition, the maximal RV end-diastolic and end-systolic cross-sectional diameters were manually measured in a basal section parallel to the tricuspid valve. To identify possible negative effects producing constriction of the RV, we evaluated diastolic filling in all patients. For this purpose, curves of global RV volume versus time were reconstructed throughout the cardiac cycle, as described previously [19]. The volume-time curves were used to calculate the following parameters of RV diastolic function, as defined by others for the LV [19]: absolute peak filling rate (PFR, the maximum slope of the diastolic filling curve), peak filling rate relative to end-diastolic volume and stroke volume of the right ventricle (PFR/EDV; PFR/SV), time to peak filling (TPF, the time from end-systole to the time of PFR), and filling fraction (FF, that percentage of the total diastolic filling that has occurred by the time of peak filling).
In addition, a subset of 5 patients underwent right heart catheterization in the usual technique before and after CSD implantation to determine possible changes in right ventricular and right atrial pressure.
Statistical analysis
The preoperative and postoperative values of the patient group were compared using Student's two-tailed t-test for paired samples. In addition, the preoperative and postoperative values of the patients were separately compared with the values of the normal controls using Student's two-tailed t-test for independent samples. All values are given as mean ± standard deviation. Significance was defined at a level of p less than 0.05.
| Results |
|---|
|
|
|---|
The results in the subgroup of 10 patients who had CSD placement alone were as follows: The NYHA class improved in all patients. Mean NYHA class was 2.7 ± 0.5 preoperatively and 1.4 ± 0.5 postoperatively.
The CT studies yielded adequate data for preoperative and postoperative evaluation. All patients had stable circulatoryfunction at the time of examination, and demonstrated sufficient opacification of the heart and no ventricular arrhythmias at the time of data acquisition.
The CT measurements revealed significant alterations in both left and right ventricular dimensions and function after CSD implantation: LV end-diastolic and end-systolic volumes decreased from 310.4 ± 87.8 mL to 232.2 ± 98.0 mL, and from 239.7 ± 78.9 mL to 159.7 ± 93.3 mL, respectively (p < 0.05 each). Whereas LV stroke volume remained nearly unchanged, the LV ejection fraction improved from 23.4% ± 6.2% to 34.4% ± 13.0% (p < 0.05).
Changes of RV parameters followed a similar pattern (Tables 1 and 2). RV end-diastolic and end-systolic volumes decreased significantly from 182.1 ± 49.6 mL to 137.5 ± 37.0 mL and from 114.8 ± 47.0 to 68.3 ± 23.8 mL, respectively. With RV stroke volume remaining nearly constant, there was a significant increase in the RV ejection fraction from 38.5 ± 8.9 to 52.0% ± 7.7%. Total RV output increased only slightly and not significantly from 5.0 ± 1.0 to 5.2 ± 1.6 L/min whereas forward RV output increased significantly from 4.0 ± 0.8 to 4.6 ± 1.1 L/min, which tends to result in a decrease in estimated tricuspid regurgitation volume from 1.0 ± 0.9 to 0.6 ± 1.0 L/min and tricuspid regurgitation fraction from 18.2% ± 14.1% to 10.4% ± 13.5%.
|
|
As presented in Table 3, all measurements reflecting early diastolic filling characteristics of the RV were nearly unchanged after surgery: no significant differences between pre- and postoperative values were found for peak filling rate, peak filling rate indexed for end-diastolic volume as well as stroke volume, for time to peak filling, and for filling fraction.
|
Postoperatively, at physical examination no patient had signs of significantly elevated RV filling pressures or constriction (such as liver congestion, ascites, or peripheral edema). These findings are confirmed by the results of invasive measurement presented in Table 4, which likewise show no tendency toward increases in RV pressures after CSD implantation.
|
| Comment |
|---|
|
|
|---|
No data are as yet available concerning the effect of CSD implantation on the RV. This is primarily due to the fact that LV parameters reflecting abnormal LV function are known to be powerful predictors of heart failure progression and survival [2426]. Moreover, RV dysfunction in dilated cardiomyopathy is often regarded as a mere consequence of the increased RV afterload resulting from LV dysfunction. Others assume that there is a direct, septum-mediated effect of the LV on the RV [27, 28]. However, the assumption that the performance of the RV merely reflects that of the LV appears to be too simple and is controversial. Thus, impaired RV contractility is also attributable to primary, intrinsic myocardial changes [29].
Furthermore, a decreased RV ejection fraction is a complementary and independent indicator of mortality and long-term outcome both in ischemic and primary dilated cardiomyopathy [1012]. A previous study has already shown that preserved RV ejection fraction at rest or during exercise is a better predictor of survival in patients with advanced heart failure than peak oxygen consumption [30].
So far, limited attention has been paid to the potential effects of CSD implantation on the structure and function of the RV with its much thinner wall compared to the LV.
To the best of our knowledge, the study presented here is the first report in the literature dealing with the effects of passive cardiac containment on the RV.
Our results indicate that CSD implantation has beneficial effects on the RV. Both end-diastolic and end-systolic volumes and diameters reflect a significant reduction in RV size. The decrease in end-diastolic and end-systolic volumes was found to be associated with a nearly unchanged total RV stroke volume, resulting in a significant increase in RV ejection fraction. With an almost constant heart rate, there was no change in total RV output while flow measurements revealed a postoperative increase in forward RV output, which is presumably due to the postoperative improvement in a concomitant relative tricuspid insufficiency, assuming that no relevant pulmonary valve regurgitation and no additional intracardiac shunt was present.
The mechanisms underlying the beneficial effects of CSD implantation on the RV have not yet been elucidated, but two mechanisms are conceivable. The first involves a decrease in the RV afterload due to a decrease in pulmonary artery pressure (resulting from improved LV function). Alternatively, there may be a direct mechanical effect of the polyester mesh device on the RV.
An indirect effect resulting from a reduced afterload is supported, for instance, by the observation that those patients who benefit from partial left ventriculectomy according to Batista by an improvement in LV function also show a better RV function but without an appreciable decrease in RV size [31].
Furthermore, when assessing RV ejection fraction, one should keep in mind that tricuspid regurgitation may be present [32]. The hemodynamic situation can thus improve when there is an increase in forward RV output despite little change in total RV output. The basis for this type of improvement is supported by results of the present study.
The direct mechanical effect of implanting a mesh device around the heart is primarily attributed to passive stabilization of ventricular geometry, the so-called "passive girdling effect" [33, 34]. With regard to the LV, the passive girdling is assumed not only to constrain ventricular dilatation but to also reduce wall tension and myocyte overextension, resulting in a decreased myocardial oxygen requirement and a persistent leftward shift of the pressure-volume relationship. This leads to cessation, or even reversal of the myocardial changes, known as reverse remodeling [4, 5, 33]. Similar effects may be operative in the RV. Finally, it should be emphasized that our study provides no evidence of right-sided constrictive physiology after device implantation. In particular, there was no indication for the assumption that diastolic function of the RV is compromised by the CSD. Postoperatively, right-sided pressure measurements showed no significant changes and none of the variables of RV diastolic filling indicated an impairment of diastolic function.
Study limitations
Our study is limited by the rather low number of patients enrolled as part of an initial nonrandomized clinical feasibility and safety study. Nevertheless changes in RV dimensions and performance reached the level of statistical significance. Moreover, it must be noted that the postoperative follow-up period of 6 to 9 months is rather short and that no randomized control group with alternative treatment of heart failure was available for comparison. The patients investigated in our study suffered from idiopathic cardiomyopathy. This is why the findings may not be readily extended to patients with ischemic cardiomyopathy.
Moreover, one has to take into account the method's systematic and random measuring errors. However, unlike other imaging modalities, electron-beam CT does not rely on simplified geometric models that may inadequately reflect the RV shape and motion. Electron-beam CT using the so-called slice summation method has been reported to be a sectional imaging modality with a high degree of accuracy and reproducibility in determining RV volumes [1316, 35].
Although magnetic resonance imaging is regarded as the gold standard for measuring the ventricular volumes, it was not suitable for the perioperative monitoring our patients due to implanted pacemakers in several cases. Therefore electron-beam CT was the method of first choice for the morphometric and functional assessment of the right ventricle in our patients and has previously been found to show good agreement with magnetic resonance imaging [36].
Conclusion
Electron-beam CT demonstrates changes in the RV geometry and function after passive cardiac containment and confirms earlier publications on the efficacy of the CSD. What remains to be elucidated is whether the observed RV improvement is due to direct mechanical effects of the CSD or to indirect hemodynamic effects of an improved LV function. Moreover, it is not clear from our data whether the changes are due to an acute effect or a progressive constrictive phenomenon induced by CSD implantation.
In summary, implantation of the CSD may be an alternative surgical treatment for patients with advanced congestive heart failure, but for a final assessment, one has to wait for the long-term results of the worldwide randomized prospective clinical trials.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Bredin, A. Olsson, and A. Franco-Cereceda No Additive Effect of Passive Containment Surgery in Patients With Aortic Regurgitation and Left Ventricular Dilation Ann. Thorac. Surg., August 1, 2007; 84(2): 510 - 513. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Chauvaud, A. C. Hernigou, E. R. Mousseaux, D. Sidi, and J.-L. Hebert Ventricular Volumes in Ebstein's Anomaly: X-Ray Multislice Computed Tomography Before and After Repair Ann. Thorac. Surg., April 1, 2006; 81(4): 1443 - 1449. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Rademakers Echocardiographic volumetry of the right ventricle Eur Heart J Cardiovasc Imaging, January 1, 2005; 6(1): 4 - 6. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |