- Email: [email protected]
Dynamic aortomyoplasty in treating end-stage heart failure
FAILING HEART—SURGICAL ASPECTS
Dynamic Aortomyoplasty in Treating End-stage Heart Failure Jorge Trainini, MD, PhD,a Edmundo I. Cabrera Fischer, MD, AFACC,b Jose´ Barisani, MD,a Alejandra I. Christen, MD,a Jorge Mouras, MD,a Jorge de Paz, MD,a Benjamı´n Elencwajg, MD,a and Juan Carlos Chachques, MDc Background: Dynamic aortomyoplasty is an alternative to heart transplantation. The goal of our study was to evaluate the benefits of aortic counterpulsation obtained using dynamic aortomyoplasty in patients with heart failure refractory to pharmacologic treatment and with contraindications to heart transplant. Methods: In this study, we compared the pre-operative and post-operative data of 15 selected patients who underwent dynamic thoracic aortomyoplasty. In this surgical technique, the right latissimus dorsi muscle flap is wrapped around the ascending aorta. This muscle flap was electrically stimulated during diastole, following a muscleconditioning protocol, to obtain diastolic augmentation. At 12-month follow-up, we evaluated invasively and non-invasively the hemodynamic, clinical, and functional effects of aortomyoplasty. Results: When comparing pre-operative data with 12-month follow-up data, we observed a significant decrease in the number of hospitalizations (p ⬍ 0.001) and in the New York Heart Association functional class (p ⬍ 0.001), and we observed significant improvement in the walking test (p ⬍ 0.001) and in peak oxygen consumption (p ⬍ 0.05). Conclusions: Dynamic, biologic, chronic counterpulsation of the aorta using a latissimus dorsi flap (dynamic aortomyoplasty) in selected patients with severe heart failure significantly improved hemodynamic parameters, heart functional data, and clinical functional class. A larger clinical experience with a longer follow-up would be useful in evaluating this technique’s clinical relevance. J Heart Lung Transplant 2002; 21:1068 –1073.
C
ongestive heart failure is an important cause of morbility and mortality in the Western world. Se-
From the aPresidente Pero ´n Hospital and bFundacio ´n Favaloro, Buenos Aires, Argentina; and cGeorges Pompidou Ho ˆpital, Paris, France. Submitted January 4, 2002; revised February 26, 2002; accepted April 3, 2002. Reprint requests: Dr. Jorge Trainini, Brandsen 1690-3 A, (1287) Buenos Aires, Argentina. Telephone and fax: 54-11-4302-3810. E-mail: [email protected] Copyright © 2002 by the International Society for Heart and Lung Transplantation. 1053-2498/02/$–see front matter PII S1053-2498(02)00438-2
1068
vere heart failure management has undergone several changes—administration of new drugs, cardiac transplantation, and ventricular assist devices—resulting in significant improvement in quality of life and survival.1–2 Nevertheless, increasing numbers of patients with cardiac diseases experience heart failure every year. According to the Framingham Heart Study, the annual survival rate for patients with heart failure was 57% for men and 64% for women, but at 5-year follow-up, survival decreased to 25% and 38%, respectively.3 Chronic cardiac failure not only is a medical
The Journal of Heart and Lung Transplantation Volume 21, Number 10
problem but also is a social and economic concern. Many patients cannot benefit from cardiac transplantation because organs are scarce and must come from healthy donors, and because of the need for immunosuppression. However, arterial counterpulsation has been used to treat acute, severe left ventricular dysfunction in patients with refractory heart failure. Indeed, arterial counterpulsation using an intra-aortic balloon pump has been used widely in acute left ventricular failure during the past 20 years. Recently, patients with refractory cardiac failure also have been treated with a biologic assistance method called dynamic aortomyoplasty.4 –7 This method consists of diastolic counterpulsation of the ascending or descending aorta through stimulation of the latissimus dorsi muscle flap wrapped around it. However, to the best of our knowledge, neither detailed clinical results nor reports of long-term follow-up have been published. The goal of this study was to evaluate hemodynamic, functional, and clinical effects of dynamic thoracic aortomyoplasty in a group of carefully selected patients who had severe heart failure refractory to pharmacologic treatment and who had contraindications for heart transplantation.
METHODS Population Study Fifteen patients (14 men) with chronic, severe heart failure were referred to Presidente Pero ´n Hospital. Mean age was 55.8 ⫾ 7.5 years (range, 44 to 69 years). All patients previously had experienced several episodes of acute pulmonary edema, despite optimal medical therapy. Four patients required inotrope therapy 1 month before aortomyoplasty. Treatment for all patients consisted of low-sodium diet, digitalis (0.25 mg/day), furosemide (40 – 120 mg/day), spironolactone (25–100 mg/day), and enalapril (5–20 mg/day). Three patients received amiodarone (200 – 400 mg/day), 6 patients received carvedilol (6.25–25 mg/day), 2 patients received nitrates (60 to 100 mg/day), and 2 patients received hydrochlorothiazide-amiloride (25 mg/day). Two patients received oral anti-coagulation therapy, and 5 patients received aspirin before surgery. Despite this personalized therapy, these patients were admitted to the hospital 4.2 ⫾ 1.6 times during the year before surgery because of congestive heart failure. Dilated cardiomyopathy resulted from Chagas disease in 4 patients, idiopathic dilated cardiomyopathy in 6 patients, and coronary artery disease in 5
Trainini et al.
1069
TABLE I Criteria for selection of patients Inclusion criteria Dilated cardiomyopathy with heart failure NYHA III–IV Left ventricular ejection fraction ⬍30% Peak VO2 ⬍16 ml/kg/min Contraindication for heart transplantation Informed patient consent Exclusion criteria Aortic regurgitation Aortic calcification Aortic aneurism or dilatation ⬎40 mm Refractory hypertension (DBP ⬎ 90 mm Hg) Arrythmia refractory to drug treatment Neuromuscular disease DBP, diastolic blood pressure; NYHA, New York Heart Association functional class; VO2, oxygen consumption.
patients. Associated cardiac pathologies included severe mitral regurgitation (n ⫽ 6), moderate mitral regurgitation (n ⫽ 6), and mild tricuspid regurgitation (n ⫽ 5). The aortic wall was explored with radioscopy and echographic studies, and no calcifications were found in these patients. Pre-operative functional class (NYHA) was 3.1 ⫾ 0.4. All patients underwent diagnostic coronary angiography; radioisotopic ventriculography and exercise thallium-201 scintigraphy detected severe stenotic lesions with no myocardium viability in 5 patients. Left ventricular ejection fraction was assessed using radioisotopic ventriculography with 99tecnesium in all patients. We based the decision for performing aortomyoplasty on previously developed criteria (Table I). All patients from this series had been rejected for cardiac transplantation because of age (n ⫽ 4), socioeconomic factors (n ⫽ 9), or alcoholism (n ⫽ 2). Severe mitral regurgitation or cardiac enlargement excluded these patients from undergoing dynamic cardiomyoplasty. Table II shows pre-operative hemodynamic and heart functional data. We obtained all data in compliance with protocols approved by the investigational review board of our institution. We informed patients of the characteristics of the procedure and its risks, and each patient gave written consent. All patients received psychologic assistance during the entire period of therapy.
Surgical Procedure In all patients, a balloon-tipped catheter recorded the right atrium pressure, pulmonary pressure, wedge pressure, and cardiac index. Afterward, with the patient in the left lateral decubitus position, a
1070
Trainini et al.
The Journal of Heart and Lung Transplantation October 2002
TABLE II Comparative analysis of pre-operative and 12-month post-operative data in patients after aortomyoplasty NYHA functional class 6-minute walking test (m) Peak VO2 (ml/kg/min) Cardiothoracic ratio LV diastolic diameter (mm) LA diameter (mm) LV shortening fraction (%) LV ejection fraction (%) PA systolic pressure (mm Hg) Wedge pressure (mm Hg) Cardiac index (liter/min/m2) Cardiac output (liter/min) Hospitalizations/p/year
Pre-AMP
Post-AMP
p
3.3 ⫾ 0.5 396 ⫾ 105 12.5 ⫾ 3.2 0.61 ⫾ 0.02 76.5 ⫾ 9.9 56.8 ⫾ 4.7 11.8 ⫾ 5.2 19.8 ⫾ 6.5 64.0 ⫾ 13.8 28.2 ⫾ 6.7 2.0 ⫾ 0.3 3.8 ⫾ 0.6 4.2 ⫾ 1.6
1.1 ⫾ 0.4 537 ⫾ 79 14.7 ⫾ 4.3 0.58 ⫾ 0.03 71.8 ⫾ 9.3 49.5 ⫾ 10.5 18.3 ⫾ 5.6 31.9 ⫾ 10.8 42.8 ⫾ 8.3 20.5 ⫾ 8.3 2.6 ⫾ 0.4 4.9 ⫾ 1.0 0.6 ⫾ 0.9
⬍0.001 ⬍0.0001 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.01 ⬍0.001 ⬍0.01 ⬍0.05 ⬍0.01 ⬍0.01 ⬍0.0001
AMP, aortomyoplasty; LA, left atrium; LV, left ventricle; NYHA, New York Heart Association; p, patient; PA, pulmonary artery; VO2, peak oxygen consumption.
wide cutaneous incision was made between the right axillary region and the posterior iliac crest. The right latissimus dorsi muscle was dissected and special care taken to preserve its neurovascular bundle. Two pacing electrodes (Medtronic 4,750, Medtronic Inc.; Minneapolis MN) were implanted into the proximal part of the latissimus dorsi muscle. The latissimus dorsi muscle flap then was stimulated to determine its contractile response (threshold, 2 V; resistance, 200 ⍀). Adequate contraction was obtained with an output of 3.5 V. Then the pedicled muscle flap was brought into the right chest after a 7-cm resection of the second right rib. After closure of this right incision, a standard median sternotomy was performed and the ascending aorta, aortic arch, and collateral vessels were dissected. When the pericardium and the right pleura were opened, the latissimus dorsi muscle flap was brought into the mediastinum and its distal portion was wrapped around the ascending aorta in a counterclockwise fashion. The observed mean external aortic diameter was 3.5 cm, the observed mean aortic length covered by the latissimus dorsi muscle flap was 6.0 cm (from aortic root to brachiocephalic trunk), and 2 muscle layers were wrapped around the aorta. Two sutureless epicardial leads (Medtronic 4,755) were fixed to the right ventricle to sense the QRS of the electrocardiogram. A cardiomyostimulator (Medtronic, model 4,710, Transform) was positioned in a left abdominal wall pocket. Both sensing and pacing leads were tunneled through sub-cutaneous tissue and connected to the cardiomyostimulator. In the last 6 patients, a minimal sternotomy (8
cm) was performed. In each of these cases, a standard pacing lead was positioned in the right ventricle through the superior vena cava.
Muscle Stimulation Protocol Skeletal muscle flap electrostimulation started 2 weeks after surgery and followed a progressive muscle-conditioning protocol for 2 months. This stimulation was similar to that performed in cardiomyoplasty but occurred during the diastolic period of the cardiac cycle. When the protocol was finished, we began stimulation: voltage, 3.5 V; pulses per burst, 6; rate mode, 1:2 with heart rate; and synchronization delay, 230 to 300 milliseconds. The delay between ventricularsensed contraction and muscle burst was adjusted through echocardiography to provide an exact synchronization between the muscle flap contraction and ventricular diastole. The burst of pulses was triggered after aortic valve closure.
Patient Follow-up At 12 months after the procedure, we evaluated invasively and non-invasively the hemodynamic and clinical effects of dynamic aortomyoplasty (Table II). We measured the extent of diastolic aortic augmentation non-invasively in 5 patients using a counterpulsation index derived from diastolic and systolic areas beneath the aortic pressure curve (DABAC/SABAC). To assess arterial pressure waves, we used a technique for non-invasive beatto-beat blood pressure monitoring in ambulatory
The Journal of Heart and Lung Transplantation Volume 21, Number 10
patients. The clinical devices used were the Finapres (2,300 Finapres, Ohmeda, CO, USA) or the Portapres (Portapres Model-2, TNO-TPD Biomedical Instrumentation; Amsterdam, The Netherlands). This technique combines the Penaz volume-clamp approach with infrared photoplethysmography, which keeps the volume of the finger artery at a pre-set value during testing. Finapres recordings of finger artery pressure waves and surface electrocardiogram were obtained during the basal state. These signals were digitized using an IBM-compatible computer. We calculated the DABAC/SABAC counterpulsation index using on-line data acquisition of the electrocardiogram and pressure signals, obtained with a Keithley DAS802 data acquisition board, driven by a specially written program in Visual Basic for Windows. Recording parameters included sampling frequency, 150 Hz; acquisition channels, 3; samples per acquisition, 800; and synchronism signal, taken from the pulse generator. We calculated the DABAC/SABAC index using a program specifically developed in C⫹⫹ language.8
Calculations and Statistical Treatment To determined the DABAC/SABAC values, we considered the association between the systolic area beneath the arterial pressure curve, enclosed between the dicrotic notch and the ascending foot of the aortic pressure wave of the following cardiac cycle, and the diastolic area beneath the arterial pressure waves, enclosed between the ascending foot and the dicrotic notch of the aortic pressure curve. Values reported are expressed as mean ⫾ standard deviation. To assess statistical significance, we used the paired Student’s t-test. We chose a 95% confidence limit to indicate statistical significance (p ⬍ 0.05).
RESULTS No intra-operative deaths occurred in this series. Inotropic pharmacologic support was routinely provided for 48 hours, and no intra-aortic counterpulsation was necessary in any case. One patient died of cardiac failure at 3 days. The other patients recovered satisfactorily from surgery, and each remained in the intensive care unit for 5 days. Post-operative hospitalization stay averaged 15.5 ⫾ 4 days. Infection of the surgical incision and of the cardiomyostimulator abdominal wall pocket developed in only 1 patient and was successfully treated with antibiotic therapy. During follow-up, 2 patients died of ventricular arrhythmia at 2 and 3 months, respectively, despite
Trainini et al.
1071
TABLE III Non-invasive evaluation of diastolic augmentation in aortomyoplasty Patient #1 Patient #2 Patient #3 Patient #4 Patient #5 Mean ⫾ SD
Non-assisted
Assisted
1.57 2.03 1.24 1.13 2.22 1.64 ⫾ 0.48
1.66 2.33 1.48 1.33 2.52 1.86 ⫾ 0.59*
Diastolic augmentation evaluated through diastolic/systolic areas beneath the arterial pressure curve relationship in assisted and unassisted cardiac cycles. Assisted vs non-assisted values showed a significant increase in diastolic augmentation in patients submitted to aortomyoplasty. *(p ⬍ 0.05)
receiving medical therapy for severe ventricular arrhythmias and atrial fibrillation. The 12 remaining patients were observed for 17.6 ⫾ 16.6 months, and currently, all are in functional NYHA Class I to II. Nine patients were observed for at least 12 months. The other 3 patients were observed for 1, 2, and 7 months, respectively. All patients received the same pharmacologic treatment after aortomyoplasty. Diuretic dosages were decreased in every patient during the postoperative period. Afterward, diuretics dosages changed throughout the follow-up of each patient. At 12-month follow-up, clinical, hemodynamic, and functional evaluation of these patients showed improvement in all parameters (Table II). Hospitalizations for congestive heart failure diminished from 4.2 ⫾ 1.6, during the year before aortomyoplasty to 0.6 ⫾ 0.9 during the first year after surgery (p ⬍ 0.001). The non-invasive counterpulsation index (DABAC/SABAC) performed in 5 patients showed significant increase between assisted and non-assisted cardiac cycles (Table III). Doppler echocardiography studies showed decreased severity of mitral and tricuspid regurgitation in all surviving patients, compared with the studies obtained before aortomyoplasty.
DISCUSSION Biologic aortic counterpulsation is a well-known technique, introduced in 1958 by Kantrowitz.9 Human application of the intra-aortic balloon pump results in beneficial hemodynamic effects, such as reduced myocardial oxygen consumption, augmentation of coronary perfusion, and increased cardiac output.10 –12 Balloon counterpulsation can be used
1072
Trainini et al.
for short- or intermediate-term circulatory assistance. During the past decade, dynamic thoracic aortomyoplasty was introduced as a new surgical technique to treat patients with severe heart failure. Because of the non-cardiac involvement of this surgical technique, it is an attractive alternative. Nature provides an example of chronic counterpulsation: diastolic augmentation is always observed in the kangaroo’s aortic pressure signal.13 Nevertheless, published clinical results and long-term follow-up with clinical, hemodynamic, and functional evaluations before and after surgery have not been detailed. Mitral regurgitation is usually found in patients with severe left ventricle dysfunction and dilatation. Bolling et al14 reported that patients with dilated cardiomyopathy and mitral regurgitation improved after mitral repair. Aortomyoplasty also decreases afterload, which improves mitral regurgitation. The Batista procedure, a partial left ventriculectomy performed in patients with dilated cardiomyopathy, continues to have a high surgical mortality of 15% and a poor life expectancy at 12-month follow-up of 56%.15 We believe that aortomyoplasty can be performed with fewer surgical risks than those of the Batista procedure. Aortomyoplasty has been performed in the ascending aorta and in the descending thoracic aorta.16 –18 We prefer the ascending aorta to the descending aorta to avoid the risk of paraplegia caused by spinal chord ischemia. By using the ascending aorta, we hoped to gain the maximal mechanical advantage of placing a counterpulsation as close as possible to the pulse origin. Diastolic augmentation of coronary blood flow also is optimized. Furthermore, the diameter of ascending aorta is larger than that of the descending aorta. The progressive muscle-conditioning protocol was designed to induce fast fatigue-resistant muscle with advantageous biomechanical characteristics for the purposes of aortomyoplasty.19 –20 We measured the total stimulation threshold, the minimum voltage needed for a total muscle contraction, during the surgical procedure and never increased stimulation beyond this threshold because high-voltage electromyostimulation has been related to muscle fibrosis. To avoid overstimulation, the stimulation rate to heart rate was 1:2. Hemodynamic and Doppler echocardiography studies demonstrated diastolic augmentation and compression of the aorta during latissimus dorsi muscle flap electrostimulation. Usually the latissimus dorsi muscle flap is used in cardiomyoplasty. Because of the enlarged ventricu-
The Journal of Heart and Lung Transplantation October 2002
lar diameters in end-stage heart failure, pericardial patches are sometimes necessary to obtain optimal cardiac wrapping.4 Nevertheless, in aortomyoplasty, the total mass of skeletal muscle required to provide an adequate counterpulsation is provided largely by the latissimus dorsi, and no supplementary patches are necessary. Our patient did not require aortic enlargement by means of a pericardial patch because the aortic root diameter was appropriate. Perhaps the volume of skeletal muscle is an important factor in the long-term integrity of pedicled flaps undergoing electrical stimulation. We also observed a decrease in left atrial and right ventricular sizes, related to decreased systolic pulmonary artery pressure and decreased cardiothoracic index. Doppler echocardiography showed a decrease in the severity of mitral and tricuspid regurgitation. Left ventricular size, measured with echocardiography, decreased, a trend, similar to the one observed with cardiomyoplasty. We also found improvement in the 6-minute walking test results, in peak oxygen consumption, in left ventricle shortening, and in ejection fraction. We observed a significant increase in the non-invasive counterpulsation index values between assisted and non-assisted cardiac cycles. This objective technique, based on the Penaz principle, may be useful in evaluating the long-term evolution of chronic counterpulsation.8 The data suggested that the procedure resulted in a decreased number of hospitalizations for congestive heart failure. In conclusion, dynamic thoracic aortomyoplasty seems to assist patients with severe heart failure refractory to drug treatment and who have contraindications for heart transplantation. Treating endstage cardiac failure with aortomyoplasty in carefully selected patients resulted in improvement of clinical, hemodynamic, and heart functional parameters. Current clinical evaluation shows improvement of functional class and quality of life for patients who undergo aortomyoplasty. A larger clinical experience, a randomized study, and a longer follow-up with periodic clinical controls and hemodynamic and functional evaluations would be useful to evaluate the long-term effects of this technique to confirm its clinical relevance. REFERENCES 1. Peters WS, Paget Milson F. Left ventricular assistance physiological, technical and clinical implications. In: Fischer EIC, Christen AI, Trainini JC, eds. Cardiovascular Failure. Bue-
The Journal of Heart and Lung Transplantation Volume 21, Number 10
2.
3.
4.
5.
6.
7.
8.
9.
10.
nos Aires: Fundacio ´n Universitaria Favaloro, 2001, pp. 287– 98. Swedberg K. Reduction in mortality by pharmacological therapy in congestive heart failure. Circulation 1993;87(suppl 4):126 –9. Ho KL, Anderson KM, Kannel WB. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993;88:107–15. Chachques JC, Grandjean PA, Cabrera Fischer EI, et al. Dynamic aortomyoplasty to assist left ventricular failure. Ann Thorac Surg 1990;49:225–30. Chachques JC, Rademercker M, Tolan MJ, Cabrera Fischer EI, Grandjean PA, Carpentier AF. Aortomyoplasty counterpulsation: experimental results and early clinical experience. Ann Thorac Surg 1996;61:420 –5. Trainini JC, Barisani JL, Elencwajg B, Cabrera Fischer E, Ahuad A, Roncoroni A. Dynamic aortomyoplasty: clinical experience. J Heart Lung Transplant 1997;16:882–4. Trainini JC, Barisani JL. Cardiac bioassist. In: Fischer EIC, Chisten AI, Trainini JC, eds. Cardiovascular Failure. Buenos Aires: Fundacio ´n Universitaria Favaloro, 2001, pp. 319 –32. Cabrera Fischer EI, Christen A, de Forteza E, Risk MR. Dynamic abdominal and thoracic aortomyoplasty in heart failure: assessment of counterpulsation. Ann Thorac Surg 1999;67:1022–9. Kantrowitz A, McKinnon WMP. The experimental use of the diaphragm as an auxiliary myocardium. Surg Forum 1959;9: 266 –8. Freed PS, Wasfie T, Zado B, Kantrowitz A. Intra-aortic balloon pumping for prolonged circulatory support. Am J Cardiol 1988;61:554 –7.
Trainini et al.
1073
11. Marchioni N, Fumagalli S, Baldereschi G, Di Bari M, Fantini F. Effective arterial pressure elastance and the hemodynamic effects of intraaortic balloon counterpulsation in patients with coronary heart disease. Am Heart J 1998;135:855–61. 12. Torchiana DF, Hirsch G, Buckley MJ, et al. Intraaortic balloon pumping for cardiac support: trends in practice and outcome 1968 to 1995. J Thorac Cardiovasc Surg 1997;13:758–69. 13. Nichols W, Avolio A, O’Rourke M. Ascending aortic impedance patterns in the kangaroo: their explanation and relation to pressure waveforms. Circ Res 1986;59:247–51. 14. Bolling SF, Deeb GM, Brunsting LA, Bach DS. Early outcome of mitral valve reconstruction in patients with end-stage cardiomyopathy. J Thorac Cardiovasc Surg 1995;109:676–83. 15. Stolf NAG, Moreira LF, Bocchi EA, et al. Determinants of midterm outcome of partial left ventriculectomy in dilated cardiomyopathy. Ann Thorac Surg 1998;66:1585–91. 16. Bolotin G, Van der Veen FH, Lorusso R, Schreuder JJ, David JB, Uretzky G. Aortomyoplasty. Basic Appl Myol 1998;8:59–65. 17. Bolotin G, Wolf T, Van der Veen F, et al. Acute descending aortomyoplasty induces coronary blood flow augmentation. Ann Thorac Surg 1999;68:1668–75. 18. Mesana T, Mouly-Bandini A, Ferzoco S, et al. Dynamic aortomyoplasty: clinical experience and thoracoscopic surgery feasibility study. J Card Surg 1998;13:60–9. 19. Salmons S, Vrbova´ G. The influence of activity on some contractile characteristics of mammalian fast and slow muscles. J Physiol 1965;201:535–49. 20. Carraro U, Chachques JC, Desnos M, Hagoge A, Fontaliran F, Carpentier A. Eight-year human dynamic cardiomyoplasty: preserved structure of myofibres and vessels of the latissimus dorsi. Basic Appl Myol 1996;6:333–6.
Dynamic Aortomyoplasty in Treating End-stage Heart Failure Jorge Trainini, MD, PhD,a Edmundo I. Cabrera Fischer, MD, AFACC,b Jose´ Barisani, MD,a Alejandra I. Christen, MD,a Jorge Mouras, MD,a Jorge de Paz, MD,a Benjamı´n Elencwajg, MD,a and Juan Carlos Chachques, MDc Background: Dynamic aortomyoplasty is an alternative to heart transplantation. The goal of our study was to evaluate the benefits of aortic counterpulsation obtained using dynamic aortomyoplasty in patients with heart failure refractory to pharmacologic treatment and with contraindications to heart transplant. Methods: In this study, we compared the pre-operative and post-operative data of 15 selected patients who underwent dynamic thoracic aortomyoplasty. In this surgical technique, the right latissimus dorsi muscle flap is wrapped around the ascending aorta. This muscle flap was electrically stimulated during diastole, following a muscleconditioning protocol, to obtain diastolic augmentation. At 12-month follow-up, we evaluated invasively and non-invasively the hemodynamic, clinical, and functional effects of aortomyoplasty. Results: When comparing pre-operative data with 12-month follow-up data, we observed a significant decrease in the number of hospitalizations (p ⬍ 0.001) and in the New York Heart Association functional class (p ⬍ 0.001), and we observed significant improvement in the walking test (p ⬍ 0.001) and in peak oxygen consumption (p ⬍ 0.05). Conclusions: Dynamic, biologic, chronic counterpulsation of the aorta using a latissimus dorsi flap (dynamic aortomyoplasty) in selected patients with severe heart failure significantly improved hemodynamic parameters, heart functional data, and clinical functional class. A larger clinical experience with a longer follow-up would be useful in evaluating this technique’s clinical relevance. J Heart Lung Transplant 2002; 21:1068 –1073.
C
ongestive heart failure is an important cause of morbility and mortality in the Western world. Se-
From the aPresidente Pero ´n Hospital and bFundacio ´n Favaloro, Buenos Aires, Argentina; and cGeorges Pompidou Ho ˆpital, Paris, France. Submitted January 4, 2002; revised February 26, 2002; accepted April 3, 2002. Reprint requests: Dr. Jorge Trainini, Brandsen 1690-3 A, (1287) Buenos Aires, Argentina. Telephone and fax: 54-11-4302-3810. E-mail: [email protected] Copyright © 2002 by the International Society for Heart and Lung Transplantation. 1053-2498/02/$–see front matter PII S1053-2498(02)00438-2
1068
vere heart failure management has undergone several changes—administration of new drugs, cardiac transplantation, and ventricular assist devices—resulting in significant improvement in quality of life and survival.1–2 Nevertheless, increasing numbers of patients with cardiac diseases experience heart failure every year. According to the Framingham Heart Study, the annual survival rate for patients with heart failure was 57% for men and 64% for women, but at 5-year follow-up, survival decreased to 25% and 38%, respectively.3 Chronic cardiac failure not only is a medical
The Journal of Heart and Lung Transplantation Volume 21, Number 10
problem but also is a social and economic concern. Many patients cannot benefit from cardiac transplantation because organs are scarce and must come from healthy donors, and because of the need for immunosuppression. However, arterial counterpulsation has been used to treat acute, severe left ventricular dysfunction in patients with refractory heart failure. Indeed, arterial counterpulsation using an intra-aortic balloon pump has been used widely in acute left ventricular failure during the past 20 years. Recently, patients with refractory cardiac failure also have been treated with a biologic assistance method called dynamic aortomyoplasty.4 –7 This method consists of diastolic counterpulsation of the ascending or descending aorta through stimulation of the latissimus dorsi muscle flap wrapped around it. However, to the best of our knowledge, neither detailed clinical results nor reports of long-term follow-up have been published. The goal of this study was to evaluate hemodynamic, functional, and clinical effects of dynamic thoracic aortomyoplasty in a group of carefully selected patients who had severe heart failure refractory to pharmacologic treatment and who had contraindications for heart transplantation.
METHODS Population Study Fifteen patients (14 men) with chronic, severe heart failure were referred to Presidente Pero ´n Hospital. Mean age was 55.8 ⫾ 7.5 years (range, 44 to 69 years). All patients previously had experienced several episodes of acute pulmonary edema, despite optimal medical therapy. Four patients required inotrope therapy 1 month before aortomyoplasty. Treatment for all patients consisted of low-sodium diet, digitalis (0.25 mg/day), furosemide (40 – 120 mg/day), spironolactone (25–100 mg/day), and enalapril (5–20 mg/day). Three patients received amiodarone (200 – 400 mg/day), 6 patients received carvedilol (6.25–25 mg/day), 2 patients received nitrates (60 to 100 mg/day), and 2 patients received hydrochlorothiazide-amiloride (25 mg/day). Two patients received oral anti-coagulation therapy, and 5 patients received aspirin before surgery. Despite this personalized therapy, these patients were admitted to the hospital 4.2 ⫾ 1.6 times during the year before surgery because of congestive heart failure. Dilated cardiomyopathy resulted from Chagas disease in 4 patients, idiopathic dilated cardiomyopathy in 6 patients, and coronary artery disease in 5
Trainini et al.
1069
TABLE I Criteria for selection of patients Inclusion criteria Dilated cardiomyopathy with heart failure NYHA III–IV Left ventricular ejection fraction ⬍30% Peak VO2 ⬍16 ml/kg/min Contraindication for heart transplantation Informed patient consent Exclusion criteria Aortic regurgitation Aortic calcification Aortic aneurism or dilatation ⬎40 mm Refractory hypertension (DBP ⬎ 90 mm Hg) Arrythmia refractory to drug treatment Neuromuscular disease DBP, diastolic blood pressure; NYHA, New York Heart Association functional class; VO2, oxygen consumption.
patients. Associated cardiac pathologies included severe mitral regurgitation (n ⫽ 6), moderate mitral regurgitation (n ⫽ 6), and mild tricuspid regurgitation (n ⫽ 5). The aortic wall was explored with radioscopy and echographic studies, and no calcifications were found in these patients. Pre-operative functional class (NYHA) was 3.1 ⫾ 0.4. All patients underwent diagnostic coronary angiography; radioisotopic ventriculography and exercise thallium-201 scintigraphy detected severe stenotic lesions with no myocardium viability in 5 patients. Left ventricular ejection fraction was assessed using radioisotopic ventriculography with 99tecnesium in all patients. We based the decision for performing aortomyoplasty on previously developed criteria (Table I). All patients from this series had been rejected for cardiac transplantation because of age (n ⫽ 4), socioeconomic factors (n ⫽ 9), or alcoholism (n ⫽ 2). Severe mitral regurgitation or cardiac enlargement excluded these patients from undergoing dynamic cardiomyoplasty. Table II shows pre-operative hemodynamic and heart functional data. We obtained all data in compliance with protocols approved by the investigational review board of our institution. We informed patients of the characteristics of the procedure and its risks, and each patient gave written consent. All patients received psychologic assistance during the entire period of therapy.
Surgical Procedure In all patients, a balloon-tipped catheter recorded the right atrium pressure, pulmonary pressure, wedge pressure, and cardiac index. Afterward, with the patient in the left lateral decubitus position, a
1070
Trainini et al.
The Journal of Heart and Lung Transplantation October 2002
TABLE II Comparative analysis of pre-operative and 12-month post-operative data in patients after aortomyoplasty NYHA functional class 6-minute walking test (m) Peak VO2 (ml/kg/min) Cardiothoracic ratio LV diastolic diameter (mm) LA diameter (mm) LV shortening fraction (%) LV ejection fraction (%) PA systolic pressure (mm Hg) Wedge pressure (mm Hg) Cardiac index (liter/min/m2) Cardiac output (liter/min) Hospitalizations/p/year
Pre-AMP
Post-AMP
p
3.3 ⫾ 0.5 396 ⫾ 105 12.5 ⫾ 3.2 0.61 ⫾ 0.02 76.5 ⫾ 9.9 56.8 ⫾ 4.7 11.8 ⫾ 5.2 19.8 ⫾ 6.5 64.0 ⫾ 13.8 28.2 ⫾ 6.7 2.0 ⫾ 0.3 3.8 ⫾ 0.6 4.2 ⫾ 1.6
1.1 ⫾ 0.4 537 ⫾ 79 14.7 ⫾ 4.3 0.58 ⫾ 0.03 71.8 ⫾ 9.3 49.5 ⫾ 10.5 18.3 ⫾ 5.6 31.9 ⫾ 10.8 42.8 ⫾ 8.3 20.5 ⫾ 8.3 2.6 ⫾ 0.4 4.9 ⫾ 1.0 0.6 ⫾ 0.9
⬍0.001 ⬍0.0001 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.01 ⬍0.001 ⬍0.01 ⬍0.05 ⬍0.01 ⬍0.01 ⬍0.0001
AMP, aortomyoplasty; LA, left atrium; LV, left ventricle; NYHA, New York Heart Association; p, patient; PA, pulmonary artery; VO2, peak oxygen consumption.
wide cutaneous incision was made between the right axillary region and the posterior iliac crest. The right latissimus dorsi muscle was dissected and special care taken to preserve its neurovascular bundle. Two pacing electrodes (Medtronic 4,750, Medtronic Inc.; Minneapolis MN) were implanted into the proximal part of the latissimus dorsi muscle. The latissimus dorsi muscle flap then was stimulated to determine its contractile response (threshold, 2 V; resistance, 200 ⍀). Adequate contraction was obtained with an output of 3.5 V. Then the pedicled muscle flap was brought into the right chest after a 7-cm resection of the second right rib. After closure of this right incision, a standard median sternotomy was performed and the ascending aorta, aortic arch, and collateral vessels were dissected. When the pericardium and the right pleura were opened, the latissimus dorsi muscle flap was brought into the mediastinum and its distal portion was wrapped around the ascending aorta in a counterclockwise fashion. The observed mean external aortic diameter was 3.5 cm, the observed mean aortic length covered by the latissimus dorsi muscle flap was 6.0 cm (from aortic root to brachiocephalic trunk), and 2 muscle layers were wrapped around the aorta. Two sutureless epicardial leads (Medtronic 4,755) were fixed to the right ventricle to sense the QRS of the electrocardiogram. A cardiomyostimulator (Medtronic, model 4,710, Transform) was positioned in a left abdominal wall pocket. Both sensing and pacing leads were tunneled through sub-cutaneous tissue and connected to the cardiomyostimulator. In the last 6 patients, a minimal sternotomy (8
cm) was performed. In each of these cases, a standard pacing lead was positioned in the right ventricle through the superior vena cava.
Muscle Stimulation Protocol Skeletal muscle flap electrostimulation started 2 weeks after surgery and followed a progressive muscle-conditioning protocol for 2 months. This stimulation was similar to that performed in cardiomyoplasty but occurred during the diastolic period of the cardiac cycle. When the protocol was finished, we began stimulation: voltage, 3.5 V; pulses per burst, 6; rate mode, 1:2 with heart rate; and synchronization delay, 230 to 300 milliseconds. The delay between ventricularsensed contraction and muscle burst was adjusted through echocardiography to provide an exact synchronization between the muscle flap contraction and ventricular diastole. The burst of pulses was triggered after aortic valve closure.
Patient Follow-up At 12 months after the procedure, we evaluated invasively and non-invasively the hemodynamic and clinical effects of dynamic aortomyoplasty (Table II). We measured the extent of diastolic aortic augmentation non-invasively in 5 patients using a counterpulsation index derived from diastolic and systolic areas beneath the aortic pressure curve (DABAC/SABAC). To assess arterial pressure waves, we used a technique for non-invasive beatto-beat blood pressure monitoring in ambulatory
The Journal of Heart and Lung Transplantation Volume 21, Number 10
patients. The clinical devices used were the Finapres (2,300 Finapres, Ohmeda, CO, USA) or the Portapres (Portapres Model-2, TNO-TPD Biomedical Instrumentation; Amsterdam, The Netherlands). This technique combines the Penaz volume-clamp approach with infrared photoplethysmography, which keeps the volume of the finger artery at a pre-set value during testing. Finapres recordings of finger artery pressure waves and surface electrocardiogram were obtained during the basal state. These signals were digitized using an IBM-compatible computer. We calculated the DABAC/SABAC counterpulsation index using on-line data acquisition of the electrocardiogram and pressure signals, obtained with a Keithley DAS802 data acquisition board, driven by a specially written program in Visual Basic for Windows. Recording parameters included sampling frequency, 150 Hz; acquisition channels, 3; samples per acquisition, 800; and synchronism signal, taken from the pulse generator. We calculated the DABAC/SABAC index using a program specifically developed in C⫹⫹ language.8
Calculations and Statistical Treatment To determined the DABAC/SABAC values, we considered the association between the systolic area beneath the arterial pressure curve, enclosed between the dicrotic notch and the ascending foot of the aortic pressure wave of the following cardiac cycle, and the diastolic area beneath the arterial pressure waves, enclosed between the ascending foot and the dicrotic notch of the aortic pressure curve. Values reported are expressed as mean ⫾ standard deviation. To assess statistical significance, we used the paired Student’s t-test. We chose a 95% confidence limit to indicate statistical significance (p ⬍ 0.05).
RESULTS No intra-operative deaths occurred in this series. Inotropic pharmacologic support was routinely provided for 48 hours, and no intra-aortic counterpulsation was necessary in any case. One patient died of cardiac failure at 3 days. The other patients recovered satisfactorily from surgery, and each remained in the intensive care unit for 5 days. Post-operative hospitalization stay averaged 15.5 ⫾ 4 days. Infection of the surgical incision and of the cardiomyostimulator abdominal wall pocket developed in only 1 patient and was successfully treated with antibiotic therapy. During follow-up, 2 patients died of ventricular arrhythmia at 2 and 3 months, respectively, despite
Trainini et al.
1071
TABLE III Non-invasive evaluation of diastolic augmentation in aortomyoplasty Patient #1 Patient #2 Patient #3 Patient #4 Patient #5 Mean ⫾ SD
Non-assisted
Assisted
1.57 2.03 1.24 1.13 2.22 1.64 ⫾ 0.48
1.66 2.33 1.48 1.33 2.52 1.86 ⫾ 0.59*
Diastolic augmentation evaluated through diastolic/systolic areas beneath the arterial pressure curve relationship in assisted and unassisted cardiac cycles. Assisted vs non-assisted values showed a significant increase in diastolic augmentation in patients submitted to aortomyoplasty. *(p ⬍ 0.05)
receiving medical therapy for severe ventricular arrhythmias and atrial fibrillation. The 12 remaining patients were observed for 17.6 ⫾ 16.6 months, and currently, all are in functional NYHA Class I to II. Nine patients were observed for at least 12 months. The other 3 patients were observed for 1, 2, and 7 months, respectively. All patients received the same pharmacologic treatment after aortomyoplasty. Diuretic dosages were decreased in every patient during the postoperative period. Afterward, diuretics dosages changed throughout the follow-up of each patient. At 12-month follow-up, clinical, hemodynamic, and functional evaluation of these patients showed improvement in all parameters (Table II). Hospitalizations for congestive heart failure diminished from 4.2 ⫾ 1.6, during the year before aortomyoplasty to 0.6 ⫾ 0.9 during the first year after surgery (p ⬍ 0.001). The non-invasive counterpulsation index (DABAC/SABAC) performed in 5 patients showed significant increase between assisted and non-assisted cardiac cycles (Table III). Doppler echocardiography studies showed decreased severity of mitral and tricuspid regurgitation in all surviving patients, compared with the studies obtained before aortomyoplasty.
DISCUSSION Biologic aortic counterpulsation is a well-known technique, introduced in 1958 by Kantrowitz.9 Human application of the intra-aortic balloon pump results in beneficial hemodynamic effects, such as reduced myocardial oxygen consumption, augmentation of coronary perfusion, and increased cardiac output.10 –12 Balloon counterpulsation can be used
1072
Trainini et al.
for short- or intermediate-term circulatory assistance. During the past decade, dynamic thoracic aortomyoplasty was introduced as a new surgical technique to treat patients with severe heart failure. Because of the non-cardiac involvement of this surgical technique, it is an attractive alternative. Nature provides an example of chronic counterpulsation: diastolic augmentation is always observed in the kangaroo’s aortic pressure signal.13 Nevertheless, published clinical results and long-term follow-up with clinical, hemodynamic, and functional evaluations before and after surgery have not been detailed. Mitral regurgitation is usually found in patients with severe left ventricle dysfunction and dilatation. Bolling et al14 reported that patients with dilated cardiomyopathy and mitral regurgitation improved after mitral repair. Aortomyoplasty also decreases afterload, which improves mitral regurgitation. The Batista procedure, a partial left ventriculectomy performed in patients with dilated cardiomyopathy, continues to have a high surgical mortality of 15% and a poor life expectancy at 12-month follow-up of 56%.15 We believe that aortomyoplasty can be performed with fewer surgical risks than those of the Batista procedure. Aortomyoplasty has been performed in the ascending aorta and in the descending thoracic aorta.16 –18 We prefer the ascending aorta to the descending aorta to avoid the risk of paraplegia caused by spinal chord ischemia. By using the ascending aorta, we hoped to gain the maximal mechanical advantage of placing a counterpulsation as close as possible to the pulse origin. Diastolic augmentation of coronary blood flow also is optimized. Furthermore, the diameter of ascending aorta is larger than that of the descending aorta. The progressive muscle-conditioning protocol was designed to induce fast fatigue-resistant muscle with advantageous biomechanical characteristics for the purposes of aortomyoplasty.19 –20 We measured the total stimulation threshold, the minimum voltage needed for a total muscle contraction, during the surgical procedure and never increased stimulation beyond this threshold because high-voltage electromyostimulation has been related to muscle fibrosis. To avoid overstimulation, the stimulation rate to heart rate was 1:2. Hemodynamic and Doppler echocardiography studies demonstrated diastolic augmentation and compression of the aorta during latissimus dorsi muscle flap electrostimulation. Usually the latissimus dorsi muscle flap is used in cardiomyoplasty. Because of the enlarged ventricu-
The Journal of Heart and Lung Transplantation October 2002
lar diameters in end-stage heart failure, pericardial patches are sometimes necessary to obtain optimal cardiac wrapping.4 Nevertheless, in aortomyoplasty, the total mass of skeletal muscle required to provide an adequate counterpulsation is provided largely by the latissimus dorsi, and no supplementary patches are necessary. Our patient did not require aortic enlargement by means of a pericardial patch because the aortic root diameter was appropriate. Perhaps the volume of skeletal muscle is an important factor in the long-term integrity of pedicled flaps undergoing electrical stimulation. We also observed a decrease in left atrial and right ventricular sizes, related to decreased systolic pulmonary artery pressure and decreased cardiothoracic index. Doppler echocardiography showed a decrease in the severity of mitral and tricuspid regurgitation. Left ventricular size, measured with echocardiography, decreased, a trend, similar to the one observed with cardiomyoplasty. We also found improvement in the 6-minute walking test results, in peak oxygen consumption, in left ventricle shortening, and in ejection fraction. We observed a significant increase in the non-invasive counterpulsation index values between assisted and non-assisted cardiac cycles. This objective technique, based on the Penaz principle, may be useful in evaluating the long-term evolution of chronic counterpulsation.8 The data suggested that the procedure resulted in a decreased number of hospitalizations for congestive heart failure. In conclusion, dynamic thoracic aortomyoplasty seems to assist patients with severe heart failure refractory to drug treatment and who have contraindications for heart transplantation. Treating endstage cardiac failure with aortomyoplasty in carefully selected patients resulted in improvement of clinical, hemodynamic, and heart functional parameters. Current clinical evaluation shows improvement of functional class and quality of life for patients who undergo aortomyoplasty. A larger clinical experience, a randomized study, and a longer follow-up with periodic clinical controls and hemodynamic and functional evaluations would be useful to evaluate the long-term effects of this technique to confirm its clinical relevance. REFERENCES 1. Peters WS, Paget Milson F. Left ventricular assistance physiological, technical and clinical implications. In: Fischer EIC, Christen AI, Trainini JC, eds. Cardiovascular Failure. Bue-
The Journal of Heart and Lung Transplantation Volume 21, Number 10
2.
3.
4.
5.
6.
7.
8.
9.
10.
nos Aires: Fundacio ´n Universitaria Favaloro, 2001, pp. 287– 98. Swedberg K. Reduction in mortality by pharmacological therapy in congestive heart failure. Circulation 1993;87(suppl 4):126 –9. Ho KL, Anderson KM, Kannel WB. Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 1993;88:107–15. Chachques JC, Grandjean PA, Cabrera Fischer EI, et al. Dynamic aortomyoplasty to assist left ventricular failure. Ann Thorac Surg 1990;49:225–30. Chachques JC, Rademercker M, Tolan MJ, Cabrera Fischer EI, Grandjean PA, Carpentier AF. Aortomyoplasty counterpulsation: experimental results and early clinical experience. Ann Thorac Surg 1996;61:420 –5. Trainini JC, Barisani JL, Elencwajg B, Cabrera Fischer E, Ahuad A, Roncoroni A. Dynamic aortomyoplasty: clinical experience. J Heart Lung Transplant 1997;16:882–4. Trainini JC, Barisani JL. Cardiac bioassist. In: Fischer EIC, Chisten AI, Trainini JC, eds. Cardiovascular Failure. Buenos Aires: Fundacio ´n Universitaria Favaloro, 2001, pp. 319 –32. Cabrera Fischer EI, Christen A, de Forteza E, Risk MR. Dynamic abdominal and thoracic aortomyoplasty in heart failure: assessment of counterpulsation. Ann Thorac Surg 1999;67:1022–9. Kantrowitz A, McKinnon WMP. The experimental use of the diaphragm as an auxiliary myocardium. Surg Forum 1959;9: 266 –8. Freed PS, Wasfie T, Zado B, Kantrowitz A. Intra-aortic balloon pumping for prolonged circulatory support. Am J Cardiol 1988;61:554 –7.
Trainini et al.
1073
11. Marchioni N, Fumagalli S, Baldereschi G, Di Bari M, Fantini F. Effective arterial pressure elastance and the hemodynamic effects of intraaortic balloon counterpulsation in patients with coronary heart disease. Am Heart J 1998;135:855–61. 12. Torchiana DF, Hirsch G, Buckley MJ, et al. Intraaortic balloon pumping for cardiac support: trends in practice and outcome 1968 to 1995. J Thorac Cardiovasc Surg 1997;13:758–69. 13. Nichols W, Avolio A, O’Rourke M. Ascending aortic impedance patterns in the kangaroo: their explanation and relation to pressure waveforms. Circ Res 1986;59:247–51. 14. Bolling SF, Deeb GM, Brunsting LA, Bach DS. Early outcome of mitral valve reconstruction in patients with end-stage cardiomyopathy. J Thorac Cardiovasc Surg 1995;109:676–83. 15. Stolf NAG, Moreira LF, Bocchi EA, et al. Determinants of midterm outcome of partial left ventriculectomy in dilated cardiomyopathy. Ann Thorac Surg 1998;66:1585–91. 16. Bolotin G, Van der Veen FH, Lorusso R, Schreuder JJ, David JB, Uretzky G. Aortomyoplasty. Basic Appl Myol 1998;8:59–65. 17. Bolotin G, Wolf T, Van der Veen F, et al. Acute descending aortomyoplasty induces coronary blood flow augmentation. Ann Thorac Surg 1999;68:1668–75. 18. Mesana T, Mouly-Bandini A, Ferzoco S, et al. Dynamic aortomyoplasty: clinical experience and thoracoscopic surgery feasibility study. J Card Surg 1998;13:60–9. 19. Salmons S, Vrbova´ G. The influence of activity on some contractile characteristics of mammalian fast and slow muscles. J Physiol 1965;201:535–49. 20. Carraro U, Chachques JC, Desnos M, Hagoge A, Fontaliran F, Carpentier A. Eight-year human dynamic cardiomyoplasty: preserved structure of myofibres and vessels of the latissimus dorsi. Basic Appl Myol 1996;6:333–6.