|
|
||||||||
Ann Thorac Surg 1998;66:1963-1968
© 1998 The Society of Thoracic Surgeons
a Dedinje Cardiovascular Institute and Belgrade University Medical School, Belgrade, Yugoslavia
Accepted for publication June 1, 1998.
Address reprint requests to Dr Gradinac, Dedinje Cardiovascular Institute, Milana Tepica 1, 11040 Belgrade, Yugoslavia
e-mail: (gradinac{at} eunet.yu)
| Abstract |
|---|
|
|
|---|
Methods. Twenty-two patients underwent partial left ventriculectomy. Mitral valve repair was performed whenever possible, otherwise the valve was replaced. Hemodynamic and functional data were obtained at baseline, as well as 2 weeks and 6 months postoperatively.
Results. Overall, 7 of 22 patients died; there were three early and four late deaths. One-year survival was 68% ± 10%. Ejection fraction increased from 23.9% ± 6.8% before the operation to 40.7% ± 12.5% at 2 weeks and to 36.8% ± 7.7% at 6 months (p < 0.001, for both). The cardiac index before the operation, at 2 weeks, and at 6 months was 2.3 ± 0.8, 2.9 ± 0.6, and 3.4 ± 1.0 L/m2 per minute, respectively (p = 0.035, and p = 0.009, compared with baseline). The increase in ejection fraction 2 weeks postoperatively was less in patients with left circumflex artery dominance (10.9% ± 3.2% compared with 19.9% ± 10.7%, respectively, p = 0.017). At 6-month follow up, all surviving patients except one improved New York Heart Association functional class when compared with preoperative status (from 3.8 ± 0.4 to 1.4 ± 0.6, p = 0.0002).
Conclusions. Early hemodynamic improvement after partial left ventriculectomy was maintained during midterm follow-up.
| Introduction |
|---|
|
|
|---|
| Material and methods |
|---|
|
|
|---|
Study patients
The final study group consisted of 22 patients with DCM who underwent PLV (20 men; mean age, 50 ± 14 years; 2 women; mean age, 51 ± 19 years). The time from the first decompensation was 16.7 ± 22.7 months, and the number of decompensations was 4.2 ± 4.0. When considered according to heart transplantation exclusion criteria [3], 6 patients would not have been accepted because they were older than 60 years, 7 patients because of high transpulmonary gradient, 1 because of previous stroke, and 1 because of hepatic failure (overall, 11 of 22, 50%). Preoperatively, NYHA class was 3.8 ± 0.4, with 7 patients (32%) receiving intravenous inotropic support. Seventeen patients (77%) had at least mild mitral regurgitation. Atrial fibrillation was present in 10 (46%) patients. Mean LV end-diastolic diameter was 79 ± 7 mm. In all patients, preoperative medical therapy included captopril, furosemide, digoxin, amiodarone, and oral anticoagulant.
Diagnostic procedures
Preoperative evaluation protocol included functional NYHA class determination, transthoracic echocardiography to measure LV end-diastolic diameter and to assess mitral regurgitation, a 6-minute walking test [11], and right and left cardiac catheterization, with coronary angiography and single-plane contrast left ventriculography, to determine LV ejection fraction (LVEF), LV volumes, and long-axis shortening. The same protocol was performed 2 weeks and 7 ± 1 months postoperatively.
Intraoperative monitoring included Swan-Ganz catheter measurements recorded at the beginning of the operation and up to 48 hours postoperatively. Transesophageal echocardiography was performed at the beginning and at the end of the operation in 20 patients to assess the short-term effects of PLV (intraoperative LVEF and mitral valve competence).
All patients had a preoperative 6-minute walking test, except those who were receiving intravenous inotropic support, to whom a 0-meter distance was ascribed as a preoperative value.
Operative procedure
High-dose glucose, insulin, and potassium metabolic support infusion was administered during the operation [12]. The originally described beating heart technique [6, 7] was used in the first 9 patients; the remaining 13 patients were operated on using crystalloid cardioplegic arrest and aortic cross-clamping.
In the beating heart technique, mild (33° to 34°C) hypothermic cardiopulmonary bypass was used. The diamond-shaped excision line of the left ventricle was narrower at the apex (Fig 1A). The attachments of both papillary muscles remained intact or, more often, were partially denuded (up to 50%), depending on interpapillary distance and the intended size of excised muscle. Midportion mitral free edge coaptation stitch was placed, using 5.0 Prolene sutures (Ethicon, Somerville, NJ) and small pledgets.
|
Transected branches of the circumflex artery were sutured with 5.0 Prolene (Ethicon) sutures after early experience with one artery stump bleeding. Ventricular suturing was done with Vicryl (Ethicon) sutures in two layers. The points of needle entry and exit were reinforced with one pass through separate felt pledgets (in the last 16 patients, Fig 1C) to prevent tears in friable muscles. When the wound was dry, two components of autogenous fibrin glue were injected consecutively between superficial layers of sutured surfaces, using an olive-tipped needle. Patients were weaned off cardiopulmonary bypass with dopamine, dobutamine, or both, and sodium nitroprusside infusions. Transesophageal echocardiography was repeated to assess LV function and mitral valve competence. Mechanical circulatory support was not needed in any patients. Autotransfusion and aprotinin were used routinely.
Follow-up
All discharged patients were evaluated monthly during a mean follow-up period of 11.4 months (range, 9 to 16 months). At each outpatient examination, NYHA functional class and 6-minute walking test distance were determined and medical therapy adjusted. Left ventriculography was not performed in 1 of 16 patents who survived 6 months, as the patient died suddenly after a 6-minute walking test, before scheduled catheterization.
Statistical analysis
Data are expressed as mean ± standard deviation. Survival curves were estimated by the Kaplan-Meier method. Hemodynamic data were compared with baseline values using a paired t test. A multiple logistic regression model was used to identify predictors of death after PLV. Variables examined included age, preoperative NYHA class, heart rate, atrial fibrillation, pulmonary capillary wedge pressure, cardiac index, mitral regurgitation grade, circumflex artery dominance, 6-minute walking test distance, LVEF, and LV end-systolic volume. A value of p less than 0.05 was considered significant.
| Results |
|---|
|
|
|---|
|
Mortality
Overall, 7 of 22 patients died during the follow-up period. Three patients died within 30 days after operation and 4 others soon thereafter. Three patients died of stroke, 2 of sepsis, and 2 had sudden death. One-year survival was 68% ± 10%. In a multiple logistic regression analysis none of the variables was predictive of death after PLV.
Early postoperative course
There were no deaths in the operating room. Thirty-day mortality was 13.7% (3 of 22). One of these patients was operated on using the beating heart technique, and two using the cardioplegic arrest technique. The first patient died of stroke on the 12th postoperative day (probable air embolism during the operation). The second death occurred in a patient with chronic atrial fibrillation who had a stroke 4 years before PLV; the patient died of a new stroke on the 16th postoperative day after an uneventful initial recovery. The third patient died of sepsis and multiorgan failure after postoperative hypertension, bleeding, rethoracotomy, renal failure, and pneumonia on the 27th postoperative day.
Follow-up
The actuarial survival curve is shown in Figure 2. Of 19 early survivors, 3 patients were readmitted to the hospital a total of five times, excluding 6-month routine check-ups. The first, a 65-year-old patient, was readmitted 2 and 4 months postoperatively, with signs of fluid retention and shortness of breath caused by noncompliance with diet and fluid restrictions. Both times he responded promptly to diuretics and fluid restrictions and his functional status returned to NYHA class II. The second patient, 62 years old, was admitted 3 months after PLV because of worsening of mitral and tricuspid regurgitation. Mitral valve replacement and tricuspid valve repair were performed. Postoperative course was complicated with orthopnea and constant cough (despite an LVEF of over 40%), which led to sternal dehiscence. The patient died of a stroke after sternal refixation, 4 months after PLV. The third readmitted patient, 59 years old, had a pacemaker implantation on the 7th postoperative day for complete atrioventricular block. Two months after PLV, he was admitted for a pacemaker wire dysfunction and injection, and has died of sepsis caused by Serratia and methicillin-resistant Staphylococcus. Two more patients in rural areas died 2 and 6 months postoperatively, after previously uneventful recovery, in NYHA class II and I. Sudden death was reported in both.
|
|
|
|
| Comment |
|---|
|
|
|---|
Our data showed that, in patients with severe LV dilation caused by nonischemic DCM, PLV resulted in a stable increase of LVEF and cardiac index. However, after an initial decrease, LV filling pressures increased toward baseline levels.
Surgical approach
Similar to other series, we modified the original technique, as the achieved correction of mitral regurgitation was inadequate [8]. The transventricular approach to the mitral valve was used to shorten ischemic time. Both additional valvuloplasty procedures and mitral valve replacement were also used as a way to reduce the mitral annulus to enable surgical reverse remodeling of basal ventricular parts. The modified technique allowed more extensive ventricular resection. The use of fibrin glue and separate pledgets for suture-line reinforcement enabled solid hold of the suture in more friable hearts and reduced the need for rethoracotomies [8]. The early improvement in LV long-axis shortening demonstrated that the procedure we used in securing the suture line did not impede LV contraction. The loss of this improvement during follow-up probably reflects the changed stress distribution of the left ventricle in the long-axis direction.
Survival
Our 30-day mortality of 14% is between the 8%, reported by Bocchi and associates [14] and 22% reported by Batista and associates [8]. However, none of the early deaths were directly caused by LV failure, which also indicates hemodynamic improvement after PLV, despite postoperative decrease in LV function. One-year mortality of 32% in the present series might reflect the learning curve, as well as the necessity to define optimal patient selection criteria and surgical strategy. It is comparable to 1-year mortality of 37% reported by Bocchi and associates [14] and 2-year mortality of 45% reported by Batista and associates [8]. On the other hand, McCarthy and associates [15] reported low 1-year mortality of 13%. However, in contrast to their data, our deaths were predominantly related to noncardiac causes, which might indicate the possibility of better survival with better perioperative and postoperative care. We did not identify any preoperative predictor of death after PLV, but the assessment of preoperative predictors of survival or sustained functional improvement probably requires a larger patient population. Also, we believe that only large, prospective, randomized study could provide data regarding specific selection criteria for PLV that would be helpful in the selection of patients who will benefit the most from this surgical procedure.
Hemodynamic improvement
Our data, as well as those of previous reports, documented early postoperative improvement of LVEF [9, 14]. Both Bocchi and associates [14] and we showed improvement of cardiac index at discharge as compared with baseline, associated with relatively high mortality. As in both series, three patients died before hemodynamic data were obtained; this might have artificially increased the postoperative cardiac index. In contrast, McCarthy and associates [15] did not find a significant increase in cardiac index intraoperatively, but the postoperative mortality was low. However they reported a significant postoperative decrease in left atrial pressure, which might be partly responsible for the inability to demonstrate an increase in cardiac index. This, as well as the small number of patients, might be the reason for the discordance in postoperative cardiac index and mortality rates in these studies. In the present study and in a study by Belotti and colleagues [16], a somewhat smaller decrease in LV filling pressures was noted, and the impact on cardiac index might have been less pronounced.
However, there is a paucity of data addressing the persistence of these effects during longer follow-up. We have shown that there is no decrease in LVEF during medium-term follow-up and that the return of LV filling pressures to baseline values is paralleled by an increase in cardiac index. A decrease in heart rate reflects better hemodynamic background.
The smaller increase in postoperative LVEF in patients with a large circumflex artery, as well as a previous report of myocardial infarctions in the ventriculotomy area [13], indicates the need to modify the original PLV procedure to minimize LV scarring. The occurrence of two sudden deaths during follow up, reported in other series as well [8, 9, 11], might also be partly attributed to the scarring process. Although all of our patients received amiodarone therapy preoperatively for at least 3 months, as suggested by a recent report [9], sudden deaths still occurred. The use of carvedilol [2] or routine electrophysiologic testing might improve outcomes [8]. In the meantime, sudden death represents the most important obstacle in making PLV widely acceptable [10].
Although preoperative LVEF and LV filling pressures are slightly better than expected, this result should be interpreted cautiously, considering that measurements were performed after the patients were optimally stabilized during the hospital stay.
One of the goals of PLV is to increase exercise capacity by improving LV systolic function. Although the present findings and those of previous studies [9, 14, 15] showed an improvement in NYHA functional class status, Bocchi and associates [14] did not find an improvement of peak oxygen consumption after PLV, which might reflect high baseline values of peak oxygen consumption in their series. We used a 6-minute walking test distance, as it was a more readily available and repeatable measure of submaximal exercise capacity and might better reflect everyday activity. Although results of this test showed a significant improvement after 6 months, longer follow-up data are needed to establish the duration of this effect.
Bolling and colleagues [17] recently demonstrated hemodynamic improvement by implantation of a flexible annuloplasty ring in heart failure patients with severe mitral regurgitation; however, only a small decrease in LV volume was achieved. Although correction of mitral regurgitation alone could not be a mechanism of improvement, it probably contributed to the hemodynamic benefit [17, 18].
As PLV enables simultaneous correction of various pathologic derangements that occur in heart failure, it might be considered as a bridge, or even an alternative to heart transplantation. Reverse ventricular remodeling achieved by PLV provides a better starting point for continued medical therapy for heart failure.
We showed that early postoperative improvements after PLV are maintained during medium-term follow-up. Partial left ventriculectomy could become an alternative for patients ineligible for transplantation or LV assist devices. Further studies are necessary to assess the effects of this operation on clinical and hemodynamic factors during long-term follow-up and to define the patient population that will benefit the most from this new surgical procedure for end-stage idiopathic dilated cardiomyopathy.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Iino, H. Ohashi, Y. Tsutsumi, T. Kawai, H. Fujii, and M. Ohnaka Biventricular Pacing 18 Months After Batista Operation Ann. Thorac. Surg., June 1, 2006; 81(6): 2302 - 2304. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Ascione, P. Wilde, and G. D. Angelini Left ventricular volume reduction MMCTS, June 28, 2005; 2005(0628): 760. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Koyama, K. Nishimura, Y. Soga, O. Unimonh, and M. Komeda Reply to the Editor J. Thorac. Cardiovasc. Surg., December 1, 2003; 126(6): 2110 - 2111. [Full Text] [PDF] |
||||
![]() |
T. Koyama, K. Nishimura, Y. Soga, O. Unimonh, K. Ueyama, and M. Komeda Importance of preserving the apex and plication of the base in left ventricular volume reduction surgery J. Thorac. Cardiovasc. Surg., March 1, 2003; 125(3): 669 - 677. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. M. Kanashiro, E. Nozawa, N. Murad, L. R. Gerola, V. A. Moises, and P. J.F. Tucci Myocardial infarction scar plication in the rat: cardiac mechanics in an animal model for surgical procedures Ann. Thorac. Surg., May 1, 2002; 73(5): 1507 - 1513. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Guccione, S. M. Moonly, A. W. Wallace, and M. B. Ratcliffe Residual stress produced by ventricular volume reduction surgery has little effect on ventricular function and mechanics: A finite element model study J. Thorac. Cardiovasc. Surg., September 1, 2001; 122(3): 592 - 599. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Bhat and R. D. Dowling Evaluation of predictors of clinical outcome after partial left ventriculectomy Ann. Thorac. Surg., July 1, 2001; 72(1): 91 - 95. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Isomura, H. Suma, T. Horii, T. Sato, T. Kobashi, H. Kanemitsu, J. Hoshino, and K. Hisatomi Left ventricle restoration in patients with non-ischemic dilated cardiomyopathy: risk factors and predictors of outcome and change of mid-term ventricular function Eur. J. Cardiothorac. Surg., May 1, 2001; 19(5): 684 - 689. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Franco-Cereceda, P. M. McCarthy, E. H. Blackstone, K. J. Hoercher, J. A. White, J. B. Young, and R. C. Starling Partial left ventriculectomy for dilated cardiomyopathy: Is this an alternative to transplantation? J. Thorac. Cardiovasc. Surg., May 1, 2001; 121(5): 879 - 893. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. F. P. Moreira, N. A.G. Stolf, M. d. L. Higuchi, F. Bacal, E. A. Bocchi, and S. A. Oliveira Current perspectives of partial left ventriculectomy in the treatment of dilated cardiomyopathy Eur. J. Cardiothorac. Surg., January 1, 2001; 19(1): 54 - 60. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Popovic, M. Miric, A.N. Neskovic, J. Vasiljevic, P. Otasevic, M. Zarkovic, M. Bojic, and S. Gradinac Functional capacity late after partial left ventriculectomy: relation to ventricular geometry and performance Eur. J. Cardiothorac. Surg., January 1, 2001; 19(1): 61 - 67. [Abstract] [Full Text] [PDF] |
||||
![]() |
O.H. Frazier, S. Gradinac, A. M. Segura, P. Przybylowski, Z. Popovic, J. Vasiljevic, A. Hernandez, H. A. McAllister Jr, M. Bojic, and B. Radovancevic Partial left ventriculectomy: which patients can be expected to benefit? Ann. Thorac. Surg., June 1, 2000; 69(6): 1836 - 1841. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Suma, T. Isomura, T. Horii, T. Sato, N. Kikuchi, K. Iwahashi, and J. Hosokawa NONTRANSPLANT CARDIAC SURGERY FOR END-STAGE CARDIOMYOPATHY J. Thorac. Cardiovasc. Surg., June 1, 2000; 119(6): 1233 - 1245. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Popovic and S Gradinac Partial left ventriculectomy improves left ventricular end systolic elastance in patients with idiopathic dilated cardiomyopathy Heart, March 1, 2000; 83(3): 316 - 319. [Abstract] [Full Text] |
||||
![]() |
L. F. P. Moreira and N. A.G. Stolf Reply Ann. Thorac. Surg., May 1, 1999; 67(5): 1542 - 1542. [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 |