WEDNESDAY AFTERNOON, May 8, 1991

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

45. Rescue Warm Substrate Blood Enriched Cardioplegia for Perioperative Sudden Death

FRIEDHELM BEYERSDORF*, MARVIN M. KIRSH,

GERALD D. BUCKBERG and BRADLEY S. ALLEN*

Frankfurt, Germany, Ann Arbor Michigan and Los Angeles, California

We reported previously experimental and clinical evidence that warm substrate enriched (glutamate/aspartate) blood cardioplegia (rescue car-dioplegia) can resuscitate energy depleted hearts and showed experimentally that myocardial salvage is possible in intractably fibrillating hearts supported by CPR for two hours.

This multi-center report describes an initial experience applying this approach to 11 patients with witnessed perioperative arrest; five elective CABG and 6 with extending infarction. All were stable hemodynamically in the cath lab, operating room or ICU when sudden irreversible ventricular fibrillation developed and progressed to electromechanical arrest in 6 patients. Conventional CPR and defibrillation were unsuccessful and extracorporeal circulation was started from 22 to 112 minutes (mean 62 minutes) after arrest. The LV was vented, the aorta clamped, and 20 minutes of warm 37°C aspartate/glutamate blood cardioplegia was given at 150 cc per minute. All bypass grafts were open with good flows in patients post-CABG, and CABG was done in the two who arrested pre-operatively.

All hearts resumed normal sinus rhythm after aortic unclamping. Ten patients had complete hemodynamic recovery, did not evolve ECG infarction, and improved postoperative ejection fraction measured at one week (42-77%; mean 57%) compared to preoperative levels (17-60%; mean 48%).* One patient was revascularized incompletely and died after 3 days of progressive cardiogenic shock. Nine patients were discharged an average of 10.5 days postoperatively. One died three months later from mediastinitis. One patient died one month postoperative from a CVA, and another sustained a small CVA (mild left hand weakness).

We conclude that witnessed peri-operative arrest with intractable ventricular fibrillation should be treated aggressively by administering CPR during prompt transfer to the operating room for institution of total vented bypass and delivery of a warm substrate enriched blood cardioplegic solution. This treatment may rescue hearts thought to be damaged irreversibly and may represent a new approach to intractable witnessed cardiac arrest, provided CPR maintains satisfactory cerebral perfusion pressure.

*p<0.05

*By Invitation

46. Experience with 28 Cases of Systolic Anterior Motion after Carpentier Mitral Valve Reconstruction

EUGENE A. GROSSI*, AUBREY C. GALLOWAY*,

STEPHENS. COLVIN*, MICHAEL A. PARISH*,

TOHRU ASAI*,. AARON J. GINDEA*,

ITZHAK KRONZON* and FRANK C. SPENCER

New York, New York

Systolic anterior motion (SAM) of the mitral valve with left ventricular outflow obstruction (LVOT) following Carpentier ring annuloplasty has led some surgeons to abandon an otherwise successful mitral reconstruction or to avoid use of a rigid annuloplasty ring. In order to determine the long-term significance of SAM, a detailed study was done on a large group of patients who had Carpentier reconstruction with ring annuloplasty (CR).

Between 3/1/81 and 6/1/90, 439 consecutive patients had CR with or without a concomitant procedure at our institution. Hospital mortality was 4.8% (21/439); LVOT obstruction may have contributed to death in one of these patients with a congenitally small aortic annulus (16 mm). Postoperative echocardiography was performed on all patients except this one, and SAM was found in 28 patients (6.4%, 28/438). Two of the 28 (7.1%) had preexisting IHSS, and 17 (60.7%) had a greater than 3 cm. resection of the posterior mitral leaflet. Only eight of the 28 patients (28.6%) had a resting LVOT gradient (mean = 53 mm Hg). These eight were treated with beta-blockers and remained asymptomatic; none required reoperation. Subsequent echocardiographic studies revealed the disappearance of SAM in 13 of the 28 patients (46.4%) and absence of any LVOT gradient in seven of eight patients (87.5%).

These data demonstrate that SAM after CR with the rigid ring: 1) occurs with a low frequency, 2) is self limiting, and 3) should be managed medically.

*By Invitation

47. Ischemic Mitral Valve Disease: A Classification and Systematic Approach to Management

JAMES H. OURY, JOSEPH C. CLEVELAND*,

WILLIAM W. ANGELL and CARLOS G. DURAN

Missoula, Montana, Tampa, Florida and Riyadh, Saudi Arabia

One-hundred-sixty-one consecutive patients with coronary artery disease (CAD) and mitral valve pathology operated during the past five years were reviewed (98% follow-up). Eighty patients underwent repair and eighty-one replacement with concomitant coronary artery bypass grafting (CABG), Av # grafts = 3. An analysis of this patient group and a classification based on anatomic pathology of the mitral apparatus is shown below.

The average age of patients in this series was 69 (range 48 to 92). Tissue valves were utilized in the majority of patients when necessary because of patient age and associated medical conditions precluding anticoagulation. Operative technique consisted of meticulous myocardial protection (antegrade and/or retrograde, i.e., coronary sinus cardioplegia), in-traoperative volume loading with transesophageal echo to monitor mitral valve hemodynamics pre and post repair, and a systematic search for the presence of pathology at each level of the mitral apparatus. Flexible ring an-nuloplasty was utilized in all repairs and chordal sparing techniques in all valve replacements. This may well account for the finding of equal mortality for replacement versus repair in this subset of high risk patients. Dysfunction of repaired valves was more common (5/74, 7.0%) than primary tissue valve failure (0 patients).

This classification by specific pathologic anatomy is recommended because it provides a basis for the choice of the appropriate valve procedure, the expected operative mortality, and long-term prognosis. Choice of surgical procedure may be selected according to this classification and the patient's age, with valve repair appropriate for younger (<60 year-old) patients in classification I and II with well-defined valve pathology.

*By Invitation

48. Universal Method for Insertion of Unstented Autografts, Homografts and Xenografts

WILLIAM W. ANGELL, DENNIS PUPELLO*

and STEPHEN HIRO*

Tampa, Florida

In order of decreasing popularity 4 methods have been described for the insertion of unstented valves in the aortic position: (1) double suture line, fully scalloped, (2) intact noncoronary sinus, (3) miniroot inclusion implant with coronary osteal approximation, (4) free aortic root insertion with pedicle coronary osteal implant.

In our series of the first 150 unstented homografts implanted using the fully scalloped double suture line method, 9 patients required reoperation for aortic regurgitation unrelated to valve deterioration within the first 36 postoperative months (2.1%/patient-year). Another 12% of patients developed early diastolic murmur. These early failures are believed to have been technical in origin due to inexact orientation of the free valve resulting in leaflet malalignment and inadequate coaptation inherent in most methods of implantation of the unstented aortic valve. This problem is not observed with the use of the free aortic root as a replacement.

A method of implanting unstented valves in the aortic position has been developed which decreases the probability of early failure resulting from technical malalignment of the valve at the time of implantation. Additionally, this technique permits the surgeon to plan and execute a standard procedure for all cases regardless of the anatomy and pathology of the recipient aortic root or the type of unstented valve selected (homograft, autograft, or xenograft). The salient features of this method are: (1) longitudinal aortotomy to the aortic annulus in the midportion of the noncoronary sinus, (2) proximal interrupted suture line with the valve oriented in the anatomic position, (3) circumferential running prolene side to side approximation of the donor coronary ostea to the recipient coronary ostea, (4) untethered com-misures to all three leaflets, (5) a distal running aortic suture line, (6) nonclosure of the aortotomy.

As with insertion of the free aortic root (method #4), this technique allows the tissue valve to assume its own natural configuration with the only dimensional requirement being an appropriate fit at the proximal suture line. It also has the advantage of leaving the recipient aortic root intact.

This method was used in a consecutive series of 21 patients including 12 homografts, 7 autografts, and 2 xenografts. Intraoperative transvalvular gradients and echocardiograms were followed by repeated echocardiogram at three month intervals. Aortic insufficiency was 0 to trivial in all cases. After one year of follow-up in 12 patients, there is no more than trace aortic insufficiency by echo, no diastolic murmur, and no instance of progressive aortic insufficiency.

In conclusion, we are proposing a method of unstented aortic valve implantation which potentially eliminates early aortic insufficiency secondary to valve malalignment and additionally leaves the recipient aortic root intact.

*By Invitation

49. Durability of Porcine Valves at 15 Years in a Representative North American Population

TOM A. BURDON*, D. CRAIG MILLER,

EDWARD S. STINSON, R. SCOTT MITCHELL*,

PHILIP E. OYER, VAUGHN A. STARNES*

and NORMAN E. SHUMWAY

Stanford, California

Broad use of bioprosthetic tissue valves has been increasingly questioned as more of these valves fail over time due to leaflet fibrocalcification. One thousand six hundred and fifty adult patients (862 AYR and 788 MVR) received a first-generation porcine xenograft bioprosthesis between 1971 and 1980. Follow-up extended to over fifteen years (total cumulative follow-up = 12,012 patient years) and was 95% complete. The average ages for AYR and MYR patients were 59 and 56 years, respectively (range = 16-87 years). Operative mortality rate was 4.5 ± 0.7% (± 70% CL) for AYR and 8.0 ± 0.9% for MVR. Two hundred and fifty-two patients are currently alive with their original bioprostheses (78 ± 1%, 55 ± 2%, and 30 ± 4% of the AYR cohort and 70 ± 2%, 50 ± 2%, and 31 ± 3% of the MVR cohort after 5, 10, and 15 years, respectively). Three hundred and seventy-two patients were withdrawn alive from the analysis after successful reoperation (either replacement of the bioprosthesis or insertion of another valve in a different site). Three hundred and ninety-eight patients underwent reoperation (REDO) to replace the original bioprosthesis; REDO AYR carried a 6.5 ± 1.9% mortality risk; this figure was 6.6 ± 1.7% for REDO MVR. A valve-related morbid or fatal event occurred in 480 patients (224 AYR and 256 MVR). Currently, 13% of AYR and 50% of MVR patients are on long-term anticoagulant therapy. Actuarial rates of freedom (± 1 SEM) from valve-related morbid events were as follows:

CONCLUSIONS: The durability of these first-generation porcine valves at 15 years was satisfactory; moreover, valve-related morbidity and mortality rates were remarkably low. Most AYR patients and at least one-half of MVR patients can safely be spared from indefinite anticoagulation without excessive risk of thromboembolism. Although one-third of the patients required reoperation, the mortality associated with first REOP was less than 7%. These observations constitute a strong argument favoring the use of porcine valves in most adult patients.

ADJOURN

*By Invitation