1:30 p.m. SCIENTIFIC SESSION - Dominion Ballroom

33. Efficacy of Dynamic Cardiomyoplasty in Dilated Cardiomyopathy

K. FRANCIS LEE*, REBECCA J. DIGNAN*,

CORNELIUS L. DYKE*, JITENDRA M. PARMAR*,

THOMAS YEH, JR*, GARY BENTON*. ANWAR S. ABD-ELFATTAH*

and ANDREW S. WECHSLER

Richmond, Virginia

Although dynamic Cardiomyoplasty (CMP) has entered the phase of active clinical trials, its direct effects on the failing heart have not been thoroughly documented. The deterioration in congestive heart failure is associated with progressive enlargement of the left ventricular chamber, decrease in cardiac output, and increased work load on the heart. We tested the hypothesis that dynamic Cardiomyoplasty may produce beneficial changes in the functional mechanics of the dilated, failing left ventricle (LV).

A model of chronic dilated Cardiomyopathy was induced in seven mongrel dogs by rapid ventricular pacing (260 b.p.m.) for three to four weeks. After the induction period, dynamic Cardiomyoplasty was performed with left latissimus dorsi, which was paced synchronously with the R-waves of the EKG (Medtronics SP1005). Instrumentation included an aortic flow probe, a LV Millar pressure catheter and piezoelectric ultrasonic crystals on the LV for measurements of wall thickness, minor axis, and major axis dimensions. Data were obtained during control (CMP-OFF) vs. study conditions (CMP-ON). Differences were analyzed with paired t-tests.

[All values are represented as means ± S.E.M. PEAK SYS, peak systole; SYS, end-systole; DIAS, end-diastole; SHORT'NG, shortening; THICK'NG, thickening. Statistical analysis is by paired t-test. N.S., not significant.]

Dynamic CMP increased the forward output of the failing heart. Systolic shortening of both minor axis and major axis dimensions were increased. Improved emptying of the congested left ventricle was associated with beneficial changes in the loading conditions of the dilated LV. Preload, as determined by LV end-diastolic pressure, decreased by 16% (p < 0.05). Although skeletal muscle contraction increased the pressure development in the LV chamber, wall stress was diminished by concomittant changes in the LV dimensions. Systolic wall thickness increased by 8% (p < 0.01); minor axis dimension decreased by 2% (p < 0.01); and major axis dimension decreased by 8% (p < 0.05). Our detailed evaluation of LV chamber mechanics suggest that dynamic cardiomyoplasty may have a role in ameliorating the functional and mechanical derangements associated with progression of dilated car-diomyopathy both by augmenting cardiac performance and by diminishing determinants of myocardial oxygen consumption.

*By Invitation

34. Left Ventricular Function Changes After Cardiomyoplasty in Patients With Dilated Cardiomyopathy

ADIB D. JATENE, NOEDIR A.G. STOLF*,

LUIZ F.P. MOREIRA*, EDIMAR A. BOCCHI*,

PAULO M.P. FERNANDES* and PEDRO SEFERIAN*

São Paulo, Brazil

Stimulated skeletal muscle flaps, contracting synchronously with the heart, have been proposed to reinforce or partially replace the ventricular walls in the treatment of irreversible myocardial failure. Between May 1988 to September 1989, latissimus dorsi Cardiomyoplasty was performed in 11 patients with dilated cardiomyopathy who were in N.Y.H.A. class III or IV despite maximal medical therapy. Mean cardiothoracic index was 58 ± 2%, resting left ventricular ejection fraction (Thechnetium 99m scintigraphy) ranged from 15 to 28% and mean cardiac index (thermodilution) was 1.78 ± 0.25 lit./min. m².

The reinforcement technique was realized without cardiopulmonary bypass and using two different incisions to muscle dissection and to cardiac access. There were no operative deaths and the prophylactic use of sym-pathomimetic amines and vasodilators provided an adequate hemodynamic condition during the surgical procedure and in the immediate postoperative period. Loss of muscle flap contraction occurred in 1 patient due to latissimus dorsi ischemia and this patient died two months after the operation in congestive heart failure. The other 10 patients followed a progressive muscle stimulation protocol for a period of three months.

Two patients remain in the course of muscle conditioning period and 8 patients were followed from 4 to 18 months with an average of 10.2 months. Four of these patients are in N.Y.H.A. class I, 3 in class II and 1 in class III. Echocardiographic evaluation showed that left ventricular (LV) segmental wall shortening increased from 11.5 ± 2.8 to 17 ± 3.9% (p<0.01) at the 3rd postoperative month and was maintained at the same levels 6 months after the operation. Similarly, Technetium 99m scintigraphy documented a LV ejection fraction improvement of 31 ± 14% (p<0.04) after the muscle conditioning period. Significant increments of LV stroke index were determined by Doppler-echocardiography and by thermodilution 3 and 6 months after Cardiomyoplasty. Values of mean pulmonary wedge pressure of 24 ± 4.2, 18.1 ± 6.5 and 17.5 ± 6.9 mmHg and LV stroke work indexes of 14 ±3.5, 23.1 ±7.5 and 21.5 ±6.8 gm.m/m² were documented by cardiac catheteriza-tion at the preoperative evaluation and at the 3rd and 6th postoperative months, respectively. In addition, the three patients who had a left ventricular end diastolic diameter greater than 80mm presented a less evident variation of the left ventricular function and the presence of moderate mitral regurgitation represented also a limitation for a better hemodynamic performance in another patient.

In conclusion, cardiomyoplasty may be performed with acceptable immediate mortality and may improve left ventricular function in patients with dilated cardiomyopathy, providing a better control of congestive heart failure state. However, the benefits of this new surgical technique seems to be less important in patients with greater left ventricular compromise.

*By Invitation

35. Clinical Experience With the Novacor Ventricular Assist System

PATRICK M. MCCARTHY*, PEER M. PORTMER*,

VAUGHN A. STARNES*, H. GARETH TOBLER*,

NARA YANAN RAMASAMY* and PHILIP E. OYER

Stanford and Oakland, California

At our institution, 12 patients (pts) had a Novacor electrical left ventricular assist system (LVAS) placed as a bridge to heart transplant (Tx) before 10/89. During the pre-LVAS hospitalization (mean 13 days), all pts were receiving inotropic support for biventricular failure (4 were also on the IABP), 10 had pulmonary edema (6 required intubation), 6 developed ventricular arrhythmias (4 had cardiac arrest), 5 had liver dysfunction with coagulopathy, and 2 had renal failure requiring artificial support. The mean cardiac index (C.I.) before LVAS was 1.52. All LVAS survivors had a dramatic increase in cardiac output (mean C.I. = 3.07). In addition, the mean PA diastolic pressure decreased from 26.4 ± 9.1 mmHg pre-implant to 17.3 ± 8.1 mmHg 24 hours post LVAS (p<0.005). One pt with cardiac allograft rejection died at LVAS implantation from akinesis of the right ventricle. Two pts died of pulmonary sepsis 16 and 21 days post LVAS implantation. Five pts were able to ambulate in the hospital and ride stationary bicycles while awaiting transplantation. Nine patients (75%) were successfully transplanted after a mean of 16 (range 2 to 34) days on the LVAS. One patient died two days after Tx of presumed sepsis. Eight pts (89%) are alive one month to five years post Tx.

In the United States, 53 pts (as of 10/89) have had a Novacor LVAS implanted. Three patients are still supported, 30 were transplanted (60%), and 26 pts (87%) survived the transplant hospitalization (one died late). No instances of device failure have occurred. In conclusion, the Novacor LVAS provided effective bridging to Tx, with post Tx survival similar to results after routine Tx. The bridge to Tx experience has provided a preview of future applications of the device. Many of these critically ill patients recovered sufficiently to allow the early resumption of normal activities. Isolated left heart support with a fully implantable LVAS will be offered as an alternative to heart Tx for selected pts in the 1990's.

*By Invitation

36. Discriminate Use of Electrocautery on the Median Sternotomy: 0.16% Wound Infection

HIROSHINISHIDA*, RONALD K. GROOTERS*,

WON-PA KIM*, HOOSHANG SOLTANZADEH*,

KENT C. THIEMAN* and ROBERT F. SCHNEIDER*

Des Moines, Iowa

Sponsored by: Ralph A. Dorner, Des Moines, Iowa

Between June, 1978, and June, 1989, superficial infection and/or deep mediastinitis developed in only 5 (0.16%) of 3,118 consecutive patients. All patients studied underwent open heart procedures through a median ster-notomy and survived more than seven postoperative days. The surgical team disciplined itself to divide presternal soft tissues with a scalpel, and used elec-trocautery only for pinpoint hemostasis to preserve tissue viability. This 0.16% incidence was statistically significantly lower than 28 previously published studies (Pearson's chi-square test, p<0.05). Twenty-five predisposing factors were evaluated by Fisher's exact test: age, sex, obesity, diabetes mellitus, smoking, chronic obstructive pulmonary disease, renal failure, malignant disease, steroid usage, the length of hospitalization before surgery, preoperative New York Heart Association classification, left ventricular end-diastolic pressure, ejection fraction, operative procedure, internal mammary artery usage, emergency operation, reoperation, duration of cardiopulmonary bypass, operating time, re-exploration, intraaortic balloon pumping, inotropic agents, perioperative myocardial infarction, stroke, and prophylactic antibiotics. Among these only the operating time longer than 3 hours is related to Sternotomy infections (p = 0.02). Statistical evidence strongly suggests that discriminate use of electrocautery minimized infection risk factors and has produced the lowest reported Sternotomy infection rate to date.

*By Invitation

37. Preservation of the Aortic Valve and Aortic Root in Ascending Aortic Dissection

JAMES I. FANN*, DONALD D. GLOWER*,

J. SCOTT RANKIN, NORMAN E. SHUMWAY,

WALTER G. WOLFE and D. CRAIG MILLER

Stanford, California and Durham, North Carolina

Surgery is the optimal approach to treatment of ascending aortic dissection; however, indications for concomitant replacement of the aortic valve and/or root remain to be defined fully. The operative approach and outcome of 282 patients admitted to two institutions with type A (ascending) aortic dissection were assessed. A total of 195 (69%) patients presented with acute and 87 (31%) with chronic type A dissections. Follow-up was complete in 98% of patients and extended to 20 years (mean 4 years). One hundred and twenty-four (44%) patients underwent aortic valve procedures: 49 (41 acute and 8 chronic) had aortic valve resuspension with an operative mortality of 10% (5/49); 75 (38 acute and 37 chronic) underwent aortic valve replacement (AYR) with an operative mortality of 21% (16/75) (p = 0.14 vs. resuspension); 16 (8 acute and 8 chronic) of these AYR patients had concomitant aortic root replacement (operative mortality of 6% [1/16]). The indications for valve replacement generally included underlying (non-acute) aortic valve disease, Marfan's syndrome, and annuloaortic ectasia. Only 2 of 49 patients who initially underwent valve resuspension required late AYR at 72 and 77 months (freedom from late AYR: 100% and 80% at 5 and 10 years). Four patients (4/75) who underwent initial AYR later required re-do AYR at 27, 81, 101, and 106 months (freedom from late re-do AYR: 98% and 70% at 5 and 10 years). Two patients (2/16) with AYR and aortic root replacement required subsequent root replacement at 48 and 71 months (freedom from late aortic root replacement: 100% and 83% at 5 and 10 years). Overall late survival rates for patients with initial AYR were 68%, 40%, and 16% at 5, 10, and 15 years, respectively. Overall late survival rates for patients with resuspension were 65% and 42%, and 21% at 5, 10, and 15 years, respectively (p = 0.5 vs.AYR). Multivariate Cox model analysis revealed that advanced age (p = 0.0001), previous cardiac or aortic operation (p = 0.0003), more preoperative complications (p = 0.0005), and earlier operative date (p = 0.006) were significant independent risk factors for late death. Preservation of the aortic valve or aortic root at initial operation was not associated with increased risk of late death or late replacement of aortic valve or aortic root. Conclusion: A conservative surgical approach to patients with type A aortic dissections favoring preservation of the native aortic valve and aortic root yields satisfactory long-term results; therefore, attempts should be made to resuspend the valve and preserve the root, whenever possible, in patients with aortic regurgitation and type A dissection.

*By Invitation

38. Effect of Right Atrial Appendectomy on the Release of Atrial Natriuretic Hormone

BASSAM O. OMARI*. RONALD J. NELSON

and JOHN M. ROBERTSON*

Torrance and Los Angeles, California

Atrial Natriuretic Hormone (ANH) is released from the right atrial appendage in response to atrial distension. During open heart surgery the right atrial appendage is usually partially removed or ligated for venous cannulation.

To evaluate the effect of right atrial appendectomy on the release of ANH and its natriuretic and diuretic responses we prospectively randomized 23 patients undergoing elective coronary artery bypass surgery into two groups. There were 16 males and 7 females with a mean age of 62 ± 7 years. Group A (n = 11) had their right atrial appendage removed and Group B (n = 12) had their right atrial appendage preserved and venous cannulation was done lateral to the appendage. Both groups were comparable for sexual distribution, age, number of grafts, and extracorporeal bypass time. All patients had normal left ventricular and renal functions, and they did not receive any diuretics during the study period.

The patients were studied both preoperatively and postoperatively whereby the ANH serum levels, urine volume/min (V), fractional excretion of sodium (FENa), amount of sodium excreted (UNaV) and central venous pressure (CVP) were measured before and after volume expansion with 7.5 cc/kg of 5% albumin. There was no statistical difference in these parameters preoperatively. However, in the postoperative period Group B patients had higher ANH levels after volume expansion that Group A patients (376 ± 84 vs 184 ± 31 pg/ml)*, higher V (4.4 ± 1.0 vs 2.0 ± 0.4 cc/min)*, higher FENa (2.30 ± 0.66 vs 0.64 ± 0.19)** and higher UNaV (511.1 ± 150.9 vs 83.9 ± 22.8 mmoles/min)**. The CVP was similar both before (10 ± 1 vs 9 ± 1 cm H₂O) and after volume expansion (16 ± 1 vs 15 ± 1 cm H₂O) in both groups.

We conclude that preserving the right atrial appendage during open heart surgery significantly increases the release of ANH, resulting in increased urinary sodium excretion and better diuresis in the post operative period. *P < 0.05 **P < 0.02.

3:30 p.m. ADJOURN

*By Invitation