Management of anesthesia during dynamic cardiomyoplasty

ORTH EINEMANN

Original Contributions Management of Anesthesia During Dynamic Cardiomyoplasty Elisa Domenegati, MD,* Marco Maurelli, MD,* Maria G. Chiaudani, MD,* Andrea Pagnin, MD,? Mauro Rinaldi, MD+ Department of Anesthesia and Intensive Care and Department of Cardiothoracic Surgery, S. Matte0 Hospital and Pavia University, 27 100 Pavia, Italy.

To review experience with anesthetic management in ten patients dynamic cardiomyoplasty (CMPL), a new surgical technique that servesas an alternative to heart transplantation. Design: Retrospective clinical study. Setting: Cardiothoracic operating room at a university hospital. Patients: Ten mule functional New York Heart Association (NYHA) classZZZand IV patients, aged 39 to 60 years, awaiting heart trans$antation, 7 of whom were diagnosed with dilated cardiomyopathy, 3 with postischemiccardiomyopathy. Interventions: Under general anesthesia, the latissimus dorsi muscle was harvested and rotated into the chest through a window in the second rib. The muscle was then wrapped around the heart. Starting from the secondpostoperative week, the lutissimus dorsi was stimulated to provide assistanceto a failing heart. Measurements and Main Results: The mean left ventricular ejection fraction (LVEF) of the 10 patients was 24.89% + 9.17% (range 10% to 37%). No intraoperative death occurred. Two patients died of multiple organ failure and an apparent arrhythmia on the 15th and 25th postoperative days, respectively. The rest of the patients regained good working capacity postoperatively, as evidenced &y improvement in NYHA grade. Nevertheless, the LVEF improved in only one patient. No significant differences were evident betweenpreoperative and postoperative blood values, hemodyn4zmicd&z, or spirometry. Conclusions: Dynamic CMPL is a considerable challenge for the anesthesiologist becausethesepatients have poor cardiac reseruepreoperatively and do not benefit from the procedure in the first two postoperative weeks. To date, CMPL seemsto be an important alternative to heart transplantation becauseexperience has shown an improvement in the quality of life with low intraoperative and postoperative complications. Study Objective:

undergoing

*Staff Anesthesiologist TAssistant Professor

of Anesthesiology

Keywords:

Anesthesia, cardiovascular;

cardiomyoplasty,

dynamic.

$Staff Surgeon Address

reprint

requests to Dr. Domene-

gati at I Servizio di Anestesia e Rianimazione, IRCCS Policlinico S. Matteo, 27100 Pavia, Italy. Received for publication January 3 1, 1994; revised manuscript accepted for publica-

tion April 19, 1994.

Introduction Although heart transplantation is an effective treatment for patients suffering from chronic heart failure refractory to pharmacologic therapy, the scarcity of donor organs greatly limits its availability. Mechanical assist devices have proven to be a useful bridge to transplantation. However, to date, no completely implantable devices have been devel-

Journal of Clinical Anesthesia 7: 177-l 8 1, 1995 0 1995 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

0952-8180/95/$10.00 SSDI 0952~8180(94)00039-7

Original Contributions

oped, and those that are currently available are associated with clinically important complications. Dynamic cardiomyoplasty (CMPL), first performed by Carpentier in Paris in 1985, is becoming an alternative treatment for the failing heart, and is being performed at an increasing number of centers.‘-’ In dynamic CMPL a skeletal muscle is wrapped around a failing heart to provide assistance with every second cardiac contraction.4 The latissimus dorsi muscle has been chosen for CMPL because it is large and easy to harvest and mobilize. Because skeletal muscle fibers differ from those in the myocardium, they must be conditioned before they become effective. A programmable, implantable cardiomyostimulator (Medtronic Inc., Minneapolis, MN) applies chronic low-frequency stimulation (at 2 Hz to 10 Hz, voltage from 3V to 8V) to the fasttwitch latissimus dorsi muscle over a period of 6 to 8 weeks. This synchronized contraction transforms the muscle into a slow-twitch muscle that is remarkably fatigue resistant. We report our experience with this procedure at the S. Matte0 Hospital in Pavia.

Materials and Methods The dynamic CMPL protocol was approved by the S. Matte0 Hospital Ethical Committee. Patients were premeditated with oral benzodiazepines one hour prior to arrival in the operating room (OR). Following placement of a large-bore intravenous (IV) cannula, a radial arterial catheter and a pulse oximeter, we used a combination of diazepam, fentanyl, and atracurium (0.5 mg/kg) for induction of anesthesia. Anesthesia was maintained with isoflurane 0.5% to 1.5%, supplemented by small boluses of fentanyl and diazepam. No further muscle relaxant boluses were administered until the threshold of the cardiomyostimulator had been tested and the contractile state of the latissimus dorsi evaluated. A peripheral nerve stimulator placed superficially along the ulnar nerve near the wrist was used to demonstrate resolution of neuromuscular block. Patients’ tracheas were intubated with a single lumen orotracheal tube and ventilated with oxygen (0,) and air (FIOz 0.5 to 0.6) by a mechanical ventilator (Siemens-Elema Servo, Solna, Sweden) in the assist-control mode. The assist capability was necessary because patients were not paralyzed. ECG, blood pressure (BP), arterial oxygen saturation by pulse oximeter (SpOa), mixed venous oxygen saturation (SvOz), and end-tidal CO* (P&O,) were continuously monitored. Inotropic drugs were available if not already continuously infused, and a cardiopulmonary bypass (CPB) machine was available if necessary. To support cardiac function during surgery, an intraaortic balloon pump was prophylactically inserted from the groin immediately after anesthesia induction. Patients were positioned on their right side for the neuromuscular phase of the procedure. After mobiliza178

J. Clin. Anesth., vol. 7, May 1995

tion of the latissimus dorsi muscle and rotation on its neurovascular pedicle into the chest through a window created by partial resection of the second rib, patients were then placed in the supine position. A further bolus of fentanyl was given prior to sternotomy and the cardiac phase of the operation was performed. During the muscle wrap, manipulation of the heart may result in ventricular arrhythmia; therefore, lidocaine 200 mg was topically applied to the myocardium. The cardiomyostimulator power pack was then inserted into a subcutaneous pocket in the abdominal wall. At the end of the operation, patients were transferred to the cardiac intensive care unit (ICU). Intravenous buprenorphine was administered every 8 hours for the first 2 days, and when needed thereafter. The muscle flap was left unstimulated for 2 weeks so that the blood supply to the distal part of the muscle could recover and the adherence of the flap to the heart surface was accomplished. At the end of the second week conditioning was begun and continued for 6 to 8 weeks. Data are expressed as mean values + standard deviation (SD) and are compared using paired Student’s t-test. A p-value less than 0.05 was considered statistically significant.

Results Since November 1992, 10 patients have undergone CMPL at our institution after written, informed consent was obtained. All patients were male aged 39 to 60 years (average 50.89, SD 7.24), in functional New York Heart Association (NYHA) class III or IV. Seven patients suffered from dilated cardiomyopathy and three from postischemic cardiomyopathy. The mean left ventricular ejection fraction (LVEF) was 24.89% (SD 9.17%, range 10 to 37%). The procedure lasted about 7.30 hours on average (range 6.30 to 9.25 hours). CPB was never needed. No intraoperative deaths occurred; 8 patients survived 1 to 14 months. Of the two patients who died, the first had low cardiac output (CO) syndrome with multiple organ failure and died on the fifteenth postoperative day in the cardiac ICU. The second patient was already discharged and found dead at home, probably from arrhythmia, on the twenty-fifth postoperative day. No patient was transfused in the OR. However, one patient received 6 units of packed red blood cells (PRBCs) and 4 units of fresh frozen plasma (FFP) in the ICU. Two patients received 4 units of PRBCs and 4 units of FFP. Another patient received 2 units of PRBCs. Furosemide was administered to 6 patients to maintain adequate urine output. Before CMPL, all patients had retired from their jobs because of their physical condition. They all had had to stop the treadmill test because of inappropriate BP response or arrhythmia, excessive fatigue and dyspnea or physical exhaustion. However, since the operation,

Anesthesiu for dynamic cardiomyoplusty: Domenegati et al.

3 patients have returned to their jobs. The rest of the patients feel subjectively better and have improved their NYHA class to I or II. They have regained a good working ability at the treadmill test with little or no dyspnea; only leg fatigue in one patient and positive BP response. Despite the improved NYHA grade, only one patient had a LVEF improvement from 30% to 47% after surgery. All other patients had more or less the same values. Preoperative spirometry was normal in six patients, two patients had mild small airway obstruction of the little airways, and one patient had chronic obstructive pulmonary disease. As shown in Table 1, postoperative values are decreased but not significantly. One patient, who had smoked 60 cigarettes daily and suffered from chronic obstructive pulmonary disease, had a preoperative forced vital capacity (FVC) of 3.38 L (80% of predicted) and forced expiratory volume in 1 minute (FEV,) of 1.43 L (43%); his postoperative values were an FVC of 2.41 L (57%) and an FEV, of 0.93 L (28%). Blood gas values and hemodynamic data are shown in Tables 2 and 3. No significant differences were noted between the preoperative and postoperative data.

Discussion Dynamic CMPL represents an alternative to cardiac transplantation in patients with end-stage ventricular failure. It presents some advantages over cardiac transplantation because it is independent of the availability of donor organs, it does not require CPB and immunosuppression, it has a low risk of infections, and of course, no rejection phenomena can occur. However, the procedure is a particular challenge to the anesthesiologist because these patients have to start and finish the operation relying on their extremely limited cardiac function. For that reason, and after the first three cases, we decided to insert an intraaortic balloon pump into all patients undergoing CMPL. We think that increasing coronary perfusion and reducing the afterload give the heart with a very low LVEF the best chance to sustain surgery and anesthesia-related stress. The intraaortic balloon pump is removed in the early postoperative period. Thanks to this procedure, CPB was not

Table 1. Preoperative and Postoperative Forced Vital

Capacity (FVC) and Forced Expiratory Volume in One Minute (FEV,) Values Rx Preoperative Postoperative

6)

4.14 (0.66) 3.60 (0.85)

Note: Data are means + SD. NS = not statistically

significant.

FEV, 04 3.40 (0.65) 2.77 (1.06)

NS

necessary, and there were no intraoperative deaths among any NYHA grade IV patients. During manipulation of the heart, premature ventricular contractions and episodes of ventricular tachycardia are frequent. We decided to administer lidocaine directly into the pericardium to prevent arrhythmias. Thus, none of the nine patients required defibrillation for sustained ventricular tachycardia or fibrillation. Only a few additional bolus doses of lidocaine were needed. One-lung ventilation was proposed by some authors, 5,6 but other researchers, such as Benasson and Kieffer’ and Greenhalgh et a1.8 disagreed. We also did not consider it necessary to collapse one lung because the latissimus dorsi muscle is a superficial one and its placement into the chest can be easily aided by manual ventilation of the lungs. Pulmonary dysfunction represents a limitation for CMPL and some institutions do not accept patients with preexisting pulmonary diseases.6 In our experience, the spirometry values (Table 1) are decreased after dissection and transposition of the latissimus dorsi muscle, but there is no statistical significance and, moreover, no clinical symptomatology. The patient with chronic obstructive pulmonary disease had a postoperative FVC of 57% of predicted and a FEVl of 28% that dramatically decreased compared to the baseline valves. It seems reasonable that before CMPL, his respiratory condition combined with his cardiac pathology ranked him in NYHA class IV. After CMPL, despite the impairment of pulmonary function tests, the patient feels better and his clinical condition has improved to a NYHA II grade, probably because he is accustomed to high degrees of airway obstruction. If not already continuously infused, an inotropic drug is prepared before induction because these patients have low cardiac reserve and are unable to compensate for the loss of sympathetic tone occurring at the induction of anesthesia, sometimes even if induction is performed very slowly. Inotropic support also may be needed following placement of the latissimus dorsi flap. We chose dobutamine as the first-line inotrope, because of its less arrhythmogenic property and enoximone plus norepinephrine in three cases of preexisting high pulmonary artery pressure. Moreover, the inotropic drugs improve blood supply to the latissimus dorsi muscle, favoring muscular tissue perfusion through a peripheric vasodilator effect. As for the muscle relaxant, some authors reverse the action of a nondepolarizing drug before the pacing leads are inserted.* Other investigators discontinue the atracurium infusion 30 minutes prior to completion of the muscle dissection.8 The Pittsburgh group uses succinylcholine for intubation, with no subsequent relaxant given. We chose atracurium because it has an intermediate duration of action and the Hofmann degradation is independent of renal or hepatic function. We administered the induction dose before intubation and without a subsequent relaxant drug. The use of 0.5% to 1.5% end-tidal isoflurane has been important to provide the J. Clin. Anesth., vol. 7, May 1995

179

0rigi?& corliribdlms Table 2. Blood Gas Analysis (mP2g) Preoperative Postoperative

83.4 (11.44) 80.7 (6.72)

36.2 (3.62) 35.5 (1.92)

HCO, (mmoI/L)

0, Saturation (%)

25.9 (4.29) 24.6 (0.65)

96.3 (1.64) 96.3 (0.70)

NS

Note: Data are means + SD. NS = not statistically significant.

low level of relaxation needed during such an operation, and it has been well tolerated from the hemodynamic point of view. Neuromuscular blockade monitoring has been important to confirm the return to a complete muscle reactivity. More information is needed about further surgery with general anesthesia, because this procedure does not preclude a future transplant, should it become necessary. Greenhalgh and coworkers’ indicate that a depolarizing muscle relaxant, which induces fasciculation, could compromise CO. We think that a nondepolarizing drug also can be dangerous and induce acute pulmonary edema. To our knowledge, only an inguinal hernia repair performed with epidural anesthesia has been described .5 It is noteworthy that, during muscle conditioning, heart rate should be maintained between 80 and 110 beats per minute (bpm): more than 80 bpm for good CO and less than 110 bpm so as not to induce an increase in muscular lactic acid until the latissimus dorsi muscle becomes fatigue resistant. According to Chachques and colleagues,g*‘O dynamic CMPL appears to exert its beneficial effects in two ways: by improving ventricular contraction and limiting cardiac dilation. Our experience shows that despite improvement in NYHA class, as confirmed both by subjective patient assessment and improvement in lifestyle, the postimplant LVEF does not show a similar increase. This finding can be due to the fact that dynamic CMPL exerts its beneficial effect primarily by limiting cardiac dilation and only secondarily by improving ventricular contraction, as documented by Capouya et al. l1 They examined the effect of an unstimulated skeletal muscle wrap in an animal model and demonstrated that the muscle flap alone has an important girdling effect on the left ven-

Table 3.

Hemodynamic

Preoperative Postoperative

tricle, which could attenuate the progressive dilation and deterioration of systolic left ventricular function seen in congestive heart failure. For that reason, they suggest performing CMPL before left ventricular dysfunction so that the need for heart transplantation may be delayed or eliminated. At present, it seems that CMPL will become an important alternative to heart transplantation since much experience has shown an improvement in the quality of life, with a low postoperative mortality.” Last but not least, CMPL offers lower health care costs after the patient is discharged since endomyocardial biopsies are not necessary, and rejection and immunosuppressionrelated infections do not occur.

References 1. Carpentier A, Chachques JC: Myocardial substitution with a stimulated skeletal muscle: first successful clinical case [Letter]. Lancet 1985; l(8440): 1267. 2. Magovern JA, Furnary AP, Christlieb IY, Kao RL, Magovern GJ: Right latissimus dorsi cardiomyoplasty for left ventricular failure. Ann Thomc Surg 1992;53: 1120-2. 3. Moreira LF, Stolf NA, Bocchi EA, et al: Latissimus dorsi cardiomyoplasty in the treatment of patients with dilated cardiomyopathy. Circulation 1992;82(5 suppl):257-63. 4. Molteni L, Almada H, Ferreira R: Synchronously stimulated skeletal muscle graft for left ventricular assistance. J Thurac Cardiovmc Surg 1989;97:43%6. 5. Robinson RJ, Truong DT, Odim J, et al: A 62-year-old man is scheduled for a new cardiac surgical procedure: dynamic cardiomyoplasty. J Cardiothorac Vast An&h 1992;6:476-87. 6. Auler JO Jr, Moreira LF, de Carvalho MJ, Stolf NA, do Amaral RV, Jatene AD: Anesthetic management of patients undergoing cardiomyoplasty. Anesthesialogy1992;77:379-81. 7. Benasson D, Kieffer JP: Preoperative management and anes-

Data

mPAP (=Hg)

PCWP (mmW)

19.12 (9.5) 22.00 (14.8)

10.50 (8.7) 14.33 (8.5)

SBP (=Hg) 113.75 (15.7) 104.67 (4.0)

co (L/mill) 5.15 (1.7) 4.10 (0.2)

2.70 (0.7) 2.20 (0.3)

NS

Note: Data are means f SD. mPAP = mean pulmonary artery pressure; PCWP = pulmonary capillary wedge pressure,. SBP = systolic blood pressure; CO = cardiac output; CI = cardiac index. 180

J. Clin. Anesth., vol. 7, May 1995

Anesthesia far dynamic cardiomyofdasty: Domenegati et al. thesia. In Carpentier A, Chachques JC, Grandjean P (eds): Cardiomy$usty. Mount Kisco, NY: Futura Publishing Inc., 1991:9%103. 8. Greenhalgh DL, Barman D, Hooper TL: Anesthesia for cardiomyoplasty. Anuesthesia 1993;48:972-4. 9. Chachques JC, Grandjean P, Schwartz K, et al: Effect of latissimus dorsi dynamic cardiomyoplasty on ventricular function. Circulation 1988;78(5 Pt 2):203-16.

10. Chachques JC, Grandjean PA, Carpentier A: Latissimus dorsi dynamic cardiomyoplasty. Ann Thurac Surg 1989;47:6O(c1. 11. Capouya ER, Gerber RS, Drinkwater DC Jr, et al: Girdling effect of nonstimulated cardiomyoplasty on left ventricular function. Ann Thora Surg 1993;56:867-70. 12. Moreira LF, Bocchi EA, Stolf NA, Pileggi F, Jatene AD: Current expectations in dynamic cardiomyoplasty. Ann Thorac Surg 1993:55:299-303.

Anaesthesia for Cardiomyoplasty DL Greenhulgh, D Barman, TL Hooper Department

of Anaesthesia,

Wythenshawe

Hospital,

Manchester,

U.K.

Abstract Cardiomyoplasty is a new surgical procedure for the treatment of severe heart failure, but early mortality has been high in the preliminary experience of many centres. This report describes the anaesthetic techniques used for the first two successful cases to be performed in the UK., Reprinted

from Anuesthesia 1993;48:972-4.

J. Clin. Anesth., vol. 7, May 1995

181