MONDAY AFTERNOON, April 18, 1988

1:45 p.m. Scientific Session - Ballroom

7. Randomized Clinical Trial of Fibrin Sealant in Cardiac Surgery Patients Undergoing Resternotomy or Reoperation - A Multicenter Study

JOHN A. ROUSOU, SIDNEY LEVITSKY, LORENZO GONZALES-LA VIN,

DELOSM. COSGROVE, HI, DONALD M. MAGILLIAN,JR.,

CLARENCE S. WELDON, CLEMENT A. HIEBERT,

PHILIP J. HESS*, LYLED. JOYCE*,

JACOB BERGSLAND* and ALAN B. GAZZANIGA

Springfield, Massachusetts; Chicago, Illinois; Browns Mills,

New Jersey; Cleveland, Ohio; Detroit, Michigan; St. Louis,

Missouri; Portland, Maine; Charlotte, North Carolina;

Minneapolis, Minnesota; Buffalo, New York and

Orange, California

A multicenter prospective study was conducted in 11 institutions to test the efficacy and safety of Fibrin Sealant (FS) as a topical hemostatic agent in patients undergoing either reoperative cardiac surgery (redo) or emergency rester-notomy. Three hundred eleven patients were randomly assigned to receive either FS (Gp. A) or a conventional topical hemostatic agent (Gp. B). The time required for bleeding to stop was recorded. If the hemostatic agent was ineffective after five minutes, the alternate hemostatic therapy was used (78 of 97 Gp. B patients received FS after 5 min). FS group was additionally compared to historical matched and unmatched controls. End-points examined include for the Concurrent Study: 1) % of bleeds controlled within 5 min, 2) mortality; for FS vs Matched Historical Controls: 1) blood loss, 2) need for resternotomy, 3) hospital stay, 4) blood products received, and 5) mortality; for FS vs Non-Matched Historical Controls (redos only): 1) Resternotomy rate. Viral studies for hepatitis B (37 pts), non-A/non-B hepatitis (33 pts) and HIV (38 pts) were carried out up to 6 mos after operation.

Results
I. Concurrent Study	Group A (FS)	Group B	p values
1. Hemostasis in:
a. Redo patients	188/202(93.1%)	11/87 (12.6%)	<0.001
b. Resternotomy patients	10/12 (83.3%)	1/10 (10.0%)	<0.05
c. Total group	198/214 (92.6%)	12/97 (11.3%)	<0.001
2. Mortality	10/166 (6.1%)	18/169 (10.7%)	N.S
II. FS vs. Hist. Matched Controls	Group A (FS)	Group B	p values
1. Blood Loss (56prs) (>1499 cc/12 hr)	1/56(1.8%)	8/56 (14.3%)	<0.05
2. Restemotomy(In Redos) (88 prs)	4/88 (4.5%)	6/88 (6.8%)	N.S.
3. Hospital Stay Days (70 prs)	12	12.9	N.S.
4. Blood Product Tx (Odds Ratio) (86 prs)	0.9	1.0	N.S.
5. Mortality (88 prs)	9/88 (10.2%)	3/88 (3.4%)	N.S.
III. FS vs. Hist Non-Matched Controls (Redos)	FS Group	Non-Matched Cont	p values
Restemotomy rate	9/159 (5.6%)	30/300 (10%)	<0.05

No patient developed hepatitis and none of those tested developed HIV viremia.

Conclusions: 1) FS is highly effective in achieving local hemostasis in cardiac surgery. 2) FS may reduce blood loss and the need for resternotomy in cardiac operations.

*By Invitation

8. A Prospective, Randomized Study of the Effect of Aprotinin on Blood Loss and Blood Usage After Coronary Bypass Operations

BENJAMIN P. BIDSTRUP*, DAVID ROYSTON*,

KENNETH M. TAYLOR* and RALPH N. SAPSFORD

London, England

Sponsored by: Delos M. COSGROVE, III,

Cleveland, Ohio

Platelet dysfunction after cardiopulmonary bypass contributes to postoperative bleeding. Dipyridamole, desmopressin acetate and prostacyclin reduce blood loss post bypass but not to levels which obviate homologous blood transfusion.

The serine protease inhibitor aprotinin (Trasylol© Bayer A.G.) was evaluated in a randomized double blind trial on 80 patients undergoing primary aorto-coronary bypass grafting. 40 patients received aprotinin (group A) by intravenous infusion (to obtain initial blood levels of >150KIU/ml) and 40 received placebo (group C). Anaesthetic, perfusion (bubble oxygenator with crystalloid prime) and surgical techniques were standardized. Postoperatively, crystalloid was infused at 1ml/kgm/hr and colloid given to maintain the central venous pressure at 8-12 mm Hg. Blood was given only if the haematocrit was less than 30%. Hydroxy-ethylated starch (Hespan) was used if it was above 30%. Template bleeding times were measured preoperatively and 90 minutes postoperatively. The two groups were similar in demographic data and bypass times. Two patients in group C required reexploration for bleeding. Both had surgical causes for the haemorrhage. One other patient from group C was excluded from analysis as he required balloon pumping. Chest tube drainage was significantly reduced in group A (301ml ±18.7 vs. 571ml ±28.5 [mean±SEM] p<0.001). The total haemoglobin loss into the chest drainage differed significantly between the two groups: (12.0 g ± 2.02 vs 37.6 g ± 3.12, p < 0.001). Group A received less bank blood postoperatively with 32 of the 40 patients receiving no blood while 37 of the 39 in group C required transfusion (p<0.001). The 8 transfused patients in group A received a total of 11 units of blood compared with a total of 78 units for group C. Haemoglobin levels were similar in both groups preoperatively and on day 1 postoperatively. The haemoglobin on day 7 postoperatively was higher in group A, (13.1g/dl ± 0.23, range 10.2-15.8 vs 12.5g/dl ± 0.2, range 9.7-15.0). In group A, the bleeding time was within the normal range 90 min postoperatively while in group C it was significantly prolonged. (6.1 min vs 12.2 min [mean]). This study demonstrates that aprotinin reduces blood losses after primary coronary artery surgery to levels which allow the majority of patients not to require bank blood.

We have also studied this compound in complex open heart procedures. 22 patients having reoperations (through a previous median sternotomy) were admitted to a randomized trial. 8 of 11 who received aprotinin lost 286 (± 48) ml into the chest drains and received no homologous blood (total given 5 units). 11 patients who did not get aprotinin lost 1509 (± 388) ml and were given a total of 41 units. An additional 10 patients (all with acute infective endocarditis), 5of whom had evidence of intravascular coagulation, were operated on with mean blood loss after aprotinin of 322 (± 76) ml. 6 had no bank blood. 11 units were given to the other 4 patients.

The ability to reduce post-bypass blood loss to levels which permit the majority of procedures to be performed without bank blood will have obvious benefits in reducing the need for a valuable, scarce and potentially hazardous resource.

*By Invitation

9. Replacement of the Transverse Aortic Arch During Emergency Surgery of Type A Acute Aortic Dissection

JEANE. BACHET*, GIOVANNI TEODORI*,

BERTRAND GOUDOT*, FERNANDO DIAZ* and

DANIEL GUILMET*

Suresnes, France

Sponsored by: ALAIN CARPENTIER,

Paris, France

In type A aortic dissection, the intimal disruption is located on, or extends to the transverse arch in about 20% patients. Replacement of the arch may, then, be necessary to avoid leaving an unresected, acutely dissected aorta and to prevent bleeding, progression of aneurysm, rupture and, ultimately, reoperation or death.

From 1970 to September 1987, 119 patients were operated on for type A acute dissection. Starting in January 1977, the GRF biological glue was used in 91 patients to reinforce the dissected tissues at the suture sites. Among these, 26 patients (aged 32 to 76 years) underwent a replacement of the transverse aortic arch in addition to the replacement of the ascending aorta.

In 20 patients (Group 1), cerebral protection was achieved by Profound Hypothermia (16 to 20°C.) associated with circulatory arrest (15 to 40 minutes, mean time : 27 mn) during the distal anastomosis.

In 6 patients (Group 2), the carotid arteries were selectively perfused with cold blood (6°C.) under moderate core hypothermia (28°C.) while Car-diopulmonary bypass was discontinued (19 to 34 minutes, mean : 25 minutes), to allow suturing of the prosthesis without crossclamping the distal aorta. Moderate hypothermia avoided the long rewarming time necessitated by profound hypothermia.

Hospital mortality accounted for 34% (9 patients out of 26). In group 1, two patients died during operation and seven patients died from postoperative complications. In group 2, no death and no major complication were observed.

Follow-up of the 17 survivors range from 3 to 90 months (mean:39). One patient died six months after surgery from cerebral hemorrhage. One patient is disabled by neurologic sequellae. Fifteen patients are in good clinical condition (NYHA Class I or II).

Postoperative aortograms in 12 patients, and CT scans in all, have shown a good and stable repair of the transverse arch in all survivors, but a persisting dissection of the descending aorta in 11 (70%).

Growing experience and improving results in emergency surgery of type A aortic dissection have led us to extend the replacement of the aorta to the transverse arch, whenever necessary. The GRF glue has proved to be an efficient adjunct. The best cerebral protection was obtained in our experience with carotid perfusion using cold blood, during circulatory arrest at moderate core hypothermia.

*By Invitation

10. Total Cavopulmonary Connection: A Logical Alternative to Atriopulmonary Connection for Complex Fontan Operations - Experimental Studies and Early Clinical Experience

MARCR. DeLEVAL*, CATHERINE BULL*

and PHILIP KILNER*

London, England

Sponsored by: D WIGHT C. McGOON

Rochester, Minnesota

Atriopulmonary connections (APC) for complex anomalies other than tricuspid atresia (complex Fontans) have disadvantages: (i) Extensive atrial surgery sometimes in the atrioventricular (AV) node area with septation and/or AV valve closure, (ii) Patients (pts) are often left with a thin walled distended right atrium (RA) prone to early and late arrhythmias and atrial thrombosis, (iii) In vitro analogues of APC demonstrate that laminar flow becomes turbulent as it enters the RA with energy loss that is not compensated for by RA contraction. Consequently, we have replaced APC for complex Fontans with a total cavopulmonary connection (TCPC). This consists of an end to side anastomosis of the superior vena cava (SVC) to the undivided right pulmonary artery (PA), the construction of a composite intra atrial tunnel using the posterior wall of the RA and a prosthetic patch to channel the inferior vena cava to the enlarged orifice of the transected SVC that is anastomosed to the main PA. The operation was performed on 13 consecutive patients aged 3-14 years, between February and October 1987. The diagnoses were double inlet ventricle (7), hypoplastic ventricle (4), transposition with straddling tricuspid valve (2). Four pts had a left SVC and underwent a bilateral SVC to PA anastomosis. There was 1 death in a pt with distorted PAs from previous banding. TCPC has the following advantages: (i) Technically simple and reproducible in any AV arrangement and away from AV node, (ii) Most of the RA chamber remains at low pressure thus reducing the risks of early and late arrhythmias, (iii) In vitro studies on casts of normal hearts on which TCPC had been performed suggest that turbulence could be alleviated thus preventing energy loss and minimizing the risks of atrial thrombosis, (iv) Postoperative cardiac catheterisation performed in 9 pts confirmed these favourable flow patterns with low caval pressures, no gradients and short transit times through the cavo pulmonary pathway. These encouraging early results support the continuing use of TCPC for complex Fontans.

*By Invitation

11. The Results of a Surgical Program for Interrupted Aortic Arch

JEFFREYE. SELL*, RICHARD A. JONAS*,

JOHN E. MA YER*, EUGENE H. BLACKSTONE,

JOHN W. KIRKLIN and ALDO R. CASTANEDA

Boston, Massachusetts and Birmingham, Alabama

Between Jan. 1, 1974 and 1987, seventy-one patients with interrupted aortic arch were admitted. Median age was 4 days, and 75% were 20 days of age or less. 63 underwent prompt initial or single stage repair; the 8 patients who died without operation within a few hours of admission or were unoperated because of extremely complex associated anomalies, are included in the data analysis. The interruption was type A in 20 (12 deaths), B in 49 (25 deaths) and C in 2 (2 deaths). An isolated ventricular septal defect (VSD) was present in 44 patients (19 deaths) and were multiple in 2 of these (2 deaths). Truncus arteriosus was present in 7 (6 deaths), and double outlet right ventricle in 5 (3 deaths). 14 patients (10 deaths) had other cardiac anomalies, and one patient had none expect bilateral patent ductuses.

Survivorship after entry (n = 71) at 1 week, 1 month, and 1, 5 and 10 years was 72%, 60%, 48%, 44% and 42% respectively (all patients but 2 were traced in Oct. 1987). The risk factors for death after repair (multivariate analysis) were (earlier) date of operation, cardiac anomalies other than single VSD, and use of a left ventricular (LV) to aortic conduit in repairing the arch interruption. The type of interruption, the type of arch repair (expect for LV conduit) and single vs. multiple stage repair were not risk factors. The improvement throughout the 13 year experience is indicated by the predicted 1 week, 1 month, and 1, 5, and 10 year survival of 96%, 94%, 91%, 89% and 89% for patients with interrupted arch and single VSD operated upon in the current era. With double inlet LV as the cardiac anomaly, these predications are 74%, 63%, 50%, 44% and 42%.

LV outflow tract obstruction (LVOTO) became evident in 12 patients. Among the 10 with VSD, the freedom from evident LVOTO at 1 week, 1 month, 1 and 3 years after VSD repair (none became evident after 3 years) was 95%, 91%, 80% and 58%. Only 1 had undergone pulmonary artery banding. The hazard function had a single phase peaking at 12 months. 8 of the 12 patients received reoperation; 11 of the 12 are alive.

Aortic reobstruction became evident in 15 patients. Freedom from this at 1 week and at 1, 5 and 10 years was 99%, 80%, 46% and 27%. Direct anastomosis was a risk factor (compared with tube graft repair) but this risk declined in recent years. Only 3 deaths occurred.

Inferences: The repair of interrupted arch with VSD by one of several methods can give good early and intermediate-term results, but follow-up is important for detection of LVOTO. In view of recent good results with single stage repair, including direct anastomosis, this technique is currently preferred. Even complex associated cardiac anomalies are not contraindications to repair, although risks are increased.

*By Invitation

12. Durability of the Viable Aortic Allograft

WILLIAM W. ANGELL, JAMES H. OURY

and JOHN J. LAMBERTI

La Jolla, California

Aortic allografts have been implanted in 503 patients since 1968. Actuarial and Hazard Function curves describe the annual incidence and freedom from valve failure in 4 groups:

1) Frozen viable allografts for AVR - 34

2) Free-sewn fresh allografts for AVR - 134

3) Stent-mounted fresh grafts for AVR - 97

4) Stent-mounted fresh grafts for MVR - 238

This series is unique in that it includes patients from the early allograft experience. All grafts were cleanly procured, antibiotic sterilized and either stored at 4 degrees centigrade for up to 8 weeks or frozen to liquid nitrogen temperatures with cryoprotection in order to preserve the viable cusp fibroblasts. In all 4 groups, valves were potentially viable at the time of implantation. Sixty-six percent of the allografts were pre-mounted on a stent to facilitate implantation and permit their use in any intracardiac position.

The longevity of this experience permits an accurate evaluation of valve durability over 10 to 20 years.

Group	Mean Time To Valve Failure		% After 10 Yrs
Frozen AVR Free	12.1 yrs. p = .05	Free 12.4 yrs.	8/11-73%
Fresh AVR Fresh	12.5 yrs. p = .05	Free 12.4 yrs.	20/36-56%
Mounted AVR	6.6 yrs. p = .0001	Mounted 8.0 yrs.	10/16-63%
Fresh Mounted MVR	8.6 yrs. p = .0001	Mounted 8.0 yrs.	14/35-40%

In all groups, viable fibroblasts were present in specimens explanted up to 5 years following surgery. All specimens returned after more than 10 years were almost totally acellular. Evidence of increased collagen suggesting that the fibroblasts survive implantations and then gradually die was present in all specimens.

The long-term durability of fresh, allovital and frozen viable allografts will determine if the complex and expensive procurement methods are justifiable. This series suggests that durability of the viable allograft for AVR is greater than for other types of tissue valves, and has an anticipated mean survival of 12 years. Pre-mounted allografts for AVR or MVR have a mean survival of 8 years and are not more durable than the glutaraldehyde porcine xenografts.

3:45 p.m. Intermission - Visit Exhibits

*By Invitation

4:30 p.m. Scientific Session - Ballroom

13. Orthotopic Heart Transplantation Survival After Total Artificial Heart Implantation

CHRISTIANE. CABROL*, IRADJ GANDJBAKHCH*,

ALAIN PAVIE*, VALERIA BORS*, TAREK MESTIRI*

ANNIK C. CABROL, EDUARDO SOLIS*,

CLA UDIO MUNERETTO* and PHILIPPE LEGER*

Paris, France

Sponsored by: NORMAN E. SHUMWAY

Stanford, California

Between April 1986 to October 1987, 24 patients underwent orthotopic implantation of a TAH (JARVIK 7). They were 21 men and 3 women with a mean age of 37,6 ± 11,3 years. From them, 11 (46%) had an adequate support and were successfully transplanted. Mean follow-up time after transplantation is 257 ± 188 days for a total of 2833 days. The mean time in mechanical support for these patients was 12 ± 5,7 days (range 2-21). Mean age was 35,6 ± 9,8 years (range 22-25). From the 11 patients, 3 (27,3%) died during the follow-up period, one died at the 10 day post-transplant from an acute rejection with sepsis, another died of anoxic coma (20 days post-transplant), and the last one of a fulminant hepatitis and pulmonary embolism (18 days post-transplant). From the 11 patients, 4 (36,4%) developed an infection in the immediate post-transplant period. Three of these infections were successfully treated with antibiotics, the other is the patient that was previously mentioned. Three patients experienced rejection episodes, reversible in two. One patient had a Kaposi sarcoma secondary to Azathioprine treatment (476 days post-transplant) and was cured. In conclusion, 46% of the patients had an adequate mechanical support and were successfully transplanted; from these patients, 72,7% are alive and well. As compared to our series of 401 heart transplantations, the mid term outcome of these patients appears to be similar as for patients with elective orthotopic heart transplantation.

*By Invitation

14. Late Results of Cardiac Transplantation Using Cyclosporine: Concerns for the Future

ROBERTL. KORMOS*, BARTLEYP. GRIFFITH,

ROBERTL. HARDESTY, JOHNM. ARMITAGE*,

JAMES NARROD* and HENRY T. BAHNSON

Pittsburgh, Pennsylvania

Current immunosupressive protocols using Cyclosporine have virtually eliminated acute rejection as a cause of late death, but it is associated with significant hypertension and renal toxicity and has not reduced the development of coronary artery disease (CAD) in the donor heart, as previously recognized with azathioprine. Between June 1980 and July 1986, 253 patients underwent cardiac transplantation, 188 (74%) of which have survived more than one year. Total later mortality has been 15% (28 pts). Annual mortality was 5% per year in the first 2 years and 2% per year thereafter. The risk of death after the first year due to infection was 5%, CAD 3%, and miscellaneous cases 7% (malignancy, 3 pts; accidental, 4 pts; liver failure, 2 pts; acute rejection, 2 pts; unknown causes, 2 pts). Evidence of acute rejection was seen in 6% of the endomyocardial biopsies performed after one year. Alterations in immunosuppressive therapy accounted for 77% of all rejection episodes. Morbidity included hypertension (92%), renal insufficiency (average serum creatinine = 1.9 ± 1 mg/dl) and progressive CAD. Seventeen patients (9%) had a serum creatinine greater than 2.5 mg/dl and 2 patients have required chronic dialysis. The incidence of CAD was 5.8% at one year and increased to 66% at year 6 with 33% of patients having developed multi-vessel disease. Based upon autopsy studies the degree of CAD was underestimated by yearly cardiac catheterization studies. The near universal need for steroids has caused diabetes mellitus in 17 (9%), the need for cataract surgery in 7 patients (4%), and osteoporotic fractures in 12 patients (6%). The early survival following cardiac transplantation justifies its proliferation, but unless strategies are developed to deal with late problems, there exists the possibility for overwhelming the already limited supply of donor hearts due to an excessive need for late retransplantation of high risk candidates.

*By Invitation

15. Decision in the Management of Sudden Cardiac Death: Endocardial Resection or Automatic Internal Defibrillator

W. CLARK HARGROVE, III*,

FRANK E. MARCHLINSKI*, MARK E. JOSEPHSON*

and JOHN M. MILLER*

Philadelphia, Pennsylvania

Sponsored by: L. HENRY EDMUNDS, JR.

Philadelphia, Pennsylvania

Subendocardial resection (SER) and implantation of the automatic internal defibrillator (AICD) are two recommended therapies for management of sudden cardiac death. Specific criteria for the choice of procedure are not developed. We reviewed 260 patients with SER and 64 with implanted AICD. Demographic, hemodynamic, arrhythmia (VT, VF or VT/VF) and survival data were analyzed in an attempt to develop selection criteria.

	Age	EF	30-Day	Survival 1 Year	At 5 Years: Freedom From: 5 Year All Arrhythmias Sudden Death
SER
n = 260	59 ± 9	27 ± 10	85%	74%	62%	85%	92%
AICD
n = 64	56 ± 2	34 ± 14	97%	78%	62%	55%	100%
p value	NS	NS	0.02	NS	NS	0.05	NS

Coronary bypass grafting was done in 69% of SER patients versus 30% of AICD patients (p<0.001). Postoperatively more AICD patients require an-tiarrhythmic drugs than SER patients (72% vs 25%, p<0.001). Pitfalls in AICD implantation include (1) occasional inability to convert VF at postoperative testing (8%), (2) failure to sense VT slower than cut-off rate (8%), (3) inability to drive an automobile (100%), (4) continued need for antiar-rhythmics (72%) and (5) inappropriate shocks (45%). In the management of malignant ventricular arrhythmias operative survival is better with AICD. Neither long term survival nor freedom from sudden death differ between the two, but SER patients have a better life style since they require fewer drugs and have a fewer arrhythmias. Death in both groups is primarily from heart failure.

5:30 p.m. Ajourn

*By Invitation