Heterogeneous Immediate Effects of Partial Left Ventriculectomy on Cardiac Performance
Background—Partial left ventriculectomy (PLV) is a novel surgical treatment for severe heart failure consisting of resection of a large wedge of myocardium to reduce wall stress and restore the normal mass-volume ratio. Although ejection fraction (EF) has been shown to improve after PLV, few other physiological data describing its immediate effects on left ventricular (LV) performance are available.
Methods and Results—Eight patients, 58±5 years old, with severe clinical heart failure and EF of 12±3% were studied before and immediately after PLV. LV performance was assessed by the predominantly load-insensitive measures of pressure-area relations with high-fidelity pressure catheters and transesophageal automated echocardiographic measures of cross-sectional area as a surrogate for volume. LV end-diastolic volume decreased from 200±60 to 89±17 mL, EF increased from 12±3% to 41±8%, and right ventricular (RV) fractional area change increased from 24±12% to 37±16% (all P<.05 versus before). Changes in pressure-area relations were variable: end-systolic elastance, 6.5±3.4 to 4.3±2.5 mm Hg/cm2 and preload recruitable stroke work, 33±16 to 34±19 mm Hg (P=NS versus before). End-diastolic stiffness increased from 0.13±0.06 to 0.19±0.07 mm Hg/cm2 (P<.05 versus before). Improvement in LV performance was inversely correlated with semiquantitative histological assessment of myocardial fibrosis and positively correlated with nuclear enlargement and hyperchromasia, indicative of myocyte hypertrophy. No long-term follow-up data were available.
Conclusions—PLV resulted in reductions in LV volumes, increases in EF and RV ejection, but increases in LV stiffness. Estimates of LV performance revealed variable results associated with the degree of myocardial fibrosis. Further study of these effects in relation to patient outcome is warranted.
Partial left ventriculectomy has been introduced recently as a surgical treatment for patients with dilated left ventricles and severe heart failure.1 This novel approach, also referred to as LV remodeling or reduction surgery, consists of resection of a large wedge of myocardium from the posterolateral wall and is often accompanied by mitral valve replacement or repair. PLV markedly reduces LV volume, with the goal of decreasing wall stress and restoring the normal mass-volume ratio. Although improvements in EF have been reported after PLV, its immediate effects on other parameters of LV performance are unclear.2 3 Furthermore, patient outcome remains variable, and details concerning patient selection, the surgical technique itself, and long-term outcome are just beginning to emerge.1 2 3 4 5 6 7 The objectives of this study were (1) to enhance understanding of the immediate effects of PLV on LV performance in patients with severe heart failure by use of relatively load-independent measures and (2) to identify potential clinical or histological variables that may be associated with improvements in LV performance.
Eight consecutive patients with severe heart failure, 58±5 years old, referred to the Hospital Angelina Caron, Brazil, for PLV were studied. All patients gave informed consent consistent with the Institutional Policy of the Hospital Angelina Caron. All patients had been in NYHA functional class III to IV for >6 months with LV EF of 12±3% (Table 1⇓). No patients were on intravenous inotropic or mechanical support. After median sternotomy, a high-fidelity LV pressure catheter (MP-500, Millar Instruments, Inc) was inserted and occluders were placed around the superior and inferior venae cavae. A transesophageal probe and an automated border-detection ultrasound system (Sonos 2500, Hewlett-Packard, Inc) that characterizes backscatter signals as blood or tissue and calculates LV cross-sectional area on-line were used.8 9 10 After a standard echocardiographic examination, automated measures of LV cross-sectional area in the transgastric short-axis view, which have previously been validated to reflect changes in LV volume in animal models and in humans, were recorded.8 9 10 This system was interfaced with an analog-to-digital converter and a computer for data storage. Three 10-second bicaval occlusion and release maneuvers were made during end-expiratory apnea to assess pressure-area relations. This protocol was repeated after PLV was complete and patients were removed from cardiopulmonary bypass but before chest closure. All operations were performed by R.J.V.B. with patients on cardiopulmonary bypass but without cardioplegia or hypothermia. A large wedge of LV free wall was resected from base to apex. All patients also had valvular surgery (Table 1⇓). Continuous suturing of the LV was performed without felt buttressing.1 Ees was determined as the slope of end-systolic points (maximum pressure/area) for each loop by an iterative linear regression technique.9 10 11 Stroke work was estimated as the integral of the pressure-area loop and PRSW as the slope of the relation of linear stroke work versus end-diastolic area.9 10 12 Assessment of LV compliance was made by use of an exponential curve fit with end-diastolic area as a surrogate for end-diastolic volume.13 Routine M-mode measures were made from short-axis images. Volume and EF measures were made by Simpson’s rule, and RV ejection was assessed as fractional area change [(end-diastolic area minus end-systolic area) divided by end-diastolic area] from the four-chamber view. Portions of excised LV wall were frozen in liquid nitrogen, then sectioned and stained with hematoxylin-eosin and Masson’s trichrome for semiquantitative assessment of endocardial and interstitial fibrosis, inflammation, necrosis, and nuclear enlargement or hyperchromasia consistent with myocyte hypertrophy. Physiological variables were compared by a paired Student’s t test from before to after PLV. Least-squares linear regression was used to assess any associations between clinical or histological features and postoperative LV performance.
Individual data appear in Table 2⇓. Patient 5 suffered a fatal intraoperative myocardial infarction and was excluded from analysis. Consistent and significant reductions in LV dimensions and volumes occurred, associated with increases in EF and RV fractional area change from 24±12% to 37±16% (P<.05 versus before). Effects of PLV on LV performance assessed by load-independent measures were more variable. Complete pressure-area loop data sets were available from six patients (Table 2⇓). Although some patients did not demonstrate an immediate benefit, other patients had a significant improvement in Ees and PRSW (Figure⇓). The most immediate improvement in LV performance occurred in a patient with chronic aortic regurgitation who also underwent aortic valve replacement. Patients with idiopathic dilated cardiomyopathy had variable results. Significant improvement in ventriculoarterial coupling was observed, with Ea decreasing from 17±6 to 11±3 mm Hg/cm2 (P<.05).14 However, a significant increase in LV stiffness occurred (Table 2⇓). Heart rate increased from 84±7 to 109±12 bpm (P<.05). LV end-diastolic and peak-systolic pressures were unchanged at 17±6 to 22±7 mm Hg and 82±10 to 71±12 mm Hg, respectively. Cardiac outputs available in four patients were also unchanged, from 3.4±0.9 to 4.5±2.1 L/min. Histological analysis revealed variable degrees of fibrosis and myocyte hypertrophy but no cellular infiltration or necrosis, consistent with end-stage heart failure. Semiquantitative grading of fibrosis was inversely correlated with changes in Ees (r=−.59) and PRSW (r=−.95, P<.01) after surgery. Degrees of nuclear enlargement and hyperchromasia associated with hypertrophy were correlated with changes in Ees (r=.70) and PRSW (r=.88, P<.05). These data suggest that improvements in LV performance are inversely related to myocardial fibrosis, in particular with PSRW, which is a more robust measure.12 The mechanism for an association with myocyte hypertrophy is unknown. This sample size was too small to draw any firm conclusions. No other clinical or surgical variables were predictive of immediate LV performance, and no long-term patient outcome data were available.
PLV for patients with severe heart failure resulted in decreases in LV volume associated with significant increases in EF and RV ejection. Increases in EF occurred even though mitral valve replacement or repair for significant mitral regurgitation was performed in the majority of patients. When the relatively load-independent measures of pressure-area relations were examined, PLV was not consistently associated with an immediate improvement in LV performance, and an increase in end-diastolic stiffness was observed. Increases in LV performance were inversely associated with myocardial fibrosis by semiquantitative histology. Although mathematical modeling predicts increases in Ees and LV stiffness with PLV, few data from patients are available.3 15 LV performance assessed by pressure-area relations has been shown to decrease immediately after coronary bypass surgery with cardioplegia and hypothermic arrest.10 This decrease is thought to be due to ischemia-reperfusion injury and/or hypothermia, although a precise mechanism is unknown. PLV is unique because of concomitant reductions in valvular regurgitation and the use of cardiopulmonary bypass without cardioplegia or hypothermic arrest in this series, although other centers use cardioplegia and core cooling. Variability in LV performance associated with the degree of myocardial fibrosis suggests that PLV may be beneficial to patients who lack advanced structural damage, although this observation requires further study in a larger series.
The world’s largest experience with partial left ventriculectomy has been the Hospital Angelina Caron in Brazil, with >400 cases.1 4 Improvement in functional class occurs in the majority of patient survivors.1 The 30-day mortality is ≈20%, and survival is estimated to be ≈55% at 2 years.1 4 Unfortunately, the outcome of many of these patients is uncertain because of their socioeconomic problems, with no means to communicate for follow-up. Indeed, no outcome data from the patients in the present study are available. McCarthy,5 Starling et al,6 and Pashkow7 have reported preliminary outcome data after PLV in a series of 53 patients, with a 1-year mortality rate of ≈6%. However, 20% to 30% of their patients had subsequent worsening of heart failure, with 15% to 20% requiring mechanical circulatory assistance.6 7 The majority of survivors had improvement in symptoms and functional class. Bocchi et al2 reported a survival rate of ≈90% at 1 month and 60% at 6 months in 24 patients with idiopathic cardiomyopathies who underwent PLV. Although these data are preliminary, a pattern of variable outcome appears to be emerging. The variable results of the load-insensitive measures of LV performance were consistent with the variable outcome reported in other patients after PLV. No attempt could be made to associate immediate LV performance with long-term outcome in this study.
A potential hypothesis is that patients who have LV dysfunction related to overstretched myocytes, rather than structural damage, are most likely to benefit from this immediate reduction in wall stress and improvement of the mass-volume relationship by PLV.16 This hypothesis is supported by the inverse relationship of immediate improvement in LV performance with degree of fibrosis. Factors such as myocyte apoptosis may possibly predict outcome; however, this hypothesis remains to be tested.17 An important future challenge is to define clinical or morphological features that would be predictive of long-term patient outcome and aid in patient selection for this surgical treatment of severe heart failure.
A major limitation of this study is the small sample size, including patients with heterogeneous clinical diagnoses. However, patients were similar with respect to the degree of LV dysfunction and heart failure. Another limitation is that follow-up data were not available, and these immediate results may not necessarily correlate with long-term outcome. A methodological limitation is the use of LV cross-sectional area as a surrogate for LV volume. This approach has been validated in animal models and humans,8 9 10 although the alterations in LV geometry induced by PLV have not been specifically studied before. In addition, increases in Ees, which usually signify improvement in LV performance, may not have the same physiological meaning after PLV.15 Also, no attempt was made to normalize Ees or PRSW to LV volume by use of stiffness-stress relations.18 Despite these limitations, the present study extends our physiological understanding of the potential immediate effects of PLV on LV performance, and these observations invite further investigation of this novel surgical therapy.
Selected Abbreviations and Acronyms
|PLV||=||partial left ventriculectomy|
|PRSW||=||preload-recruitable stroke work|
Dr Gorcsan was supported in part by the American Heart Association, Pennsylvania Affiliate, Camp Hill.
- Received December 2, 1997.
- Revision received January 7, 1998.
- Accepted January 12, 1998.
- Copyright © 1998 by American Heart Association
Bocchi EA, Bellotti G, de Moraes AV, Bacal F, Moreia LF, Esteves-Filho A, Fukushima T, Guimaraes G, Stolf N, Jantene A, Pileggi F. Clinical outcome after left ventricular remodeling in patients with idiopathic dilated cardiomyopathy referred to heart transplantation: short-term results. Circulation. 1997;96(suppl II):II-165–II-172. Comment.
Kawaguchi AT, Sugimachi M, Sunagawa K, Takeshita N, Koide S, Verde JL, Batista RJV. Intraoperative left ventricular pressure-volume relationship in patients undergoing left ventricular diameter reduction. Circulation. 1997;96(suppl I):I-198. Abstract.
Starling RC, Young JB, Scalia GM, Thomas JD, Vargo RL, Buda TM, Smedira NG, McCarthy PM. Preliminary observations with ventricular remodeling surgery for refractory heart failure. J Am Coll Cardiol. 1997;29:64A. Abstract.
Pashkow FJ, ed. Partial left ventriculectomy a viable alternative to cardiac transplant. Cardiac Consult.. 1997;8:4.
Gorcsan J, Gasior TA, Mandarino WA, Deneault LG, Hattler BG, Pinsky MR. Assessment of the immediate effects of cardiopulmonary bypass on left ventricular performance by on-line pressure-area relations. Circulation. 1994;89:180–190.
Suga H, Sagawa K, Shoukas AA. Load independence of the instantaneous pressure-volume ratio of the canine left ventricle and effects of epinephrine and heart rate on the ratio. Circ Res. 1973;32:314–322.
Glower DD, Spratt JA, Snow ND, Kabas JS, Davis JW, Olsen CO, Tyson GS, Sabiston DC, Rankin JS. Linearity of the Frank-Starling relationship in the intact heart: the concept of preload recruitable stroke work. Circulation. 1985;5:994–1009.
Mirsky I, Cohn PF, Levine JA, Gorlin R, Herman MV, Kreulen TH, Sonnenblick EH. Assessment of left ventricular stiffness in primary myocardial and coronary artery disease. Circulation. 1974;50:128–136.
Morita S, Kormos RL, Mandarino WA, Eishi K, Kawai A, Gasior TA, Deneault LG, Armitage JM, Hardesty RL, Griffith BG. Right ventricular/arterial coupling in the patient with left ventricular assistance. Circulation. 1992;86(suppl II):II-316-II-325.
Nakano K, Sugawara M, Ishihara K, Kanazawa S, Corin WJ, Denslow S, Biederman RW, Carabello BA. Myocardial stiffness derived from end- systolic wall stress and logarithm of reciprocal of wall thickness: contractility index independent of ventricular size. Circulation. 1990;82:1352–1361.