WEDNESDAY MORNING, APRIL 30,1980

8:30 A.M. Scientific Session - Continental Ballroom

29. Clinical Durability of the (Hancock) Porcine Xenograft Valve

PHILIP E. OVER*, D. CRAIG MILLER*,

EDWARD B. STINSON*, BRUCE A REITZ*,

RICARDO MORENO-CABRAL*, and

NORMAN E. SHUMWAY, Stanford, California

The principal feature which remains to be completely defined in regard to the Hancock xenograft bioprosthesis is long-term durability. As increasing numbers of bioprosthetic valves are implanted, this characteristic assumes greater importance, particularly since some investigators have recently advocated abandonment of xenografts because of impaired function after 5 years (yrs). This report provides extended data regarding valve durability derived from 1407 patients (707 aortic [AYR] and 700 mitral [MVR] replacements) who received Hancock bioprostheses between 1971 and 1979; cumulative duration of follow-up was 1732 patient-years (pt-yrs) for AYR and 1843 pt-yrs for MVR, and maximum followup was 8.4 yrs. One hundred seventy nine pts were followed for more than 5 yrs and 67 for more than 6 yrs. Valve failure was defined on the basis of the following criteria: 1) development of a new regurgitant murmur 2) thrombotic valvular occlusion 3) infective endocarditis resulting in reoperation or death 4) hemodynamic valvular dysfunction confirmed by catheterization and resulting in reoperation or death. Using these criteria 15 valve failures occurred in the AVR cohort and 23 in the MVR cohort, yielding linearized occurrence rates of 0.9%/pt-yr and 1.3%/pt-yr., respectively. Actuarial analysis revealed that 96 ± 1.2% (± SEM) of AVR pts after 5 yrs and 90 ± 2.6% of MVR pts after 6 yrs were free of valve failure. Primary tissue failure (defined by calcification or leaflet disruption resulting in stenosis and/or regurgitation without antecedent endocarditis) or valve thrombosis accounted for 5 of 15 AVR failures and 13 of 23 MVR failures, equivalent to linearized occurrence rates of 0.3%/ pt-yr and 0.7%/ pt-yr, respectively. Calculation by actuarial methods revealed that 98 ± 1% of AVR (at 5 yrs) and 92.5 ± 2.4% of MVR pts (at 6 yrs) were free of primary tissue failure or valve thrombosis. The actuarially-calculated rate of valve failure up to 5 yrs remains constant as a function of time for AVR pts; for MVR pts there appears to be a slight acceleration in the rate of valve failure after 5 yrs. In this extended experience the incidence of xenograft valve failure 5 to 6 yrs postoperatively remains acceptably low and supports the continued use of the xenograft bioprosthesis for cardiac valve replacement.

*By invitation

30. Risk-Benefit Analysis of Warfarin Therapy in Hancock Mitral Valve Replacement

J. DONALD HILL, LIZELLEN LA FOLLETTE*.

ROBERT J. SZARNICKI*, G. JAMES A VERY, II*,

WILLIAM J. KERTH, FRANK GERBODE,

and ROBERT RODVIEN*, San Francisco, California

Porcine mitral valve bioprostheses have a low thromboembolic rate. Controversy persists concerning whether warfarin lowers this rate further without undue risk. In mid-1976 we changed from no systemic anticoag-ulation to routine systemic anticoagulation begun on day 3. The purpose, therefore, of this investigation was to analyze the thromboembolic and/or major bleeding complications of 124 consecutive but nonrandomized patients who had only Hancock mitral valve replacements between 9/74 and 6/79 treated with and without anticoagulants. Four basic study groups were created: Group 1, warfarin; Group 2, aspirin; Group 3, no anticoagulants; Group 4, warfarin and aspirin. Group 5 combines Groups 1 and 4 (warfarin and warfarin and aspirin) and Group 6 combines Groups 2 and 3 (aspirin and no antigoagulants).

Group Therapy	1 Warfarin	2 Aspirin	3 No anti- coagulants	4 Warfarin+ Aspirin	5 Combined 1 + 4	6 Combined 2 + 3
No. of Patients	72	21	26	5	77	47
Mean follow-up (yr)	2.2	4.2	3.7	2.0	2.2	3.9
Patient years	158.2	87.8	96.7	9.9	168.1	184.4
Emboli	4	2	3	0	4	5
Emboli per 100 patient years	2.5	2.3	3.1	0.0	2.4	2.7
Major bleeding complications	9	1	0.0	0.0	9	1
Major bleeding per 100 patient yrs	5.7	1.1	0.0	0.0	5.4	0.5

The embolic rate was not significantly different in any group (P = NS). There were no deaths from emboli in any group. Group 1 (warfarin) resulted in significant major bleeding complications (P<0.05), including one death. Bleeding was also increased in Group 5 (warfarin, and warfarin and aspirin) (P< 0.01).

Conclusions: There is no difference in the thromboembolic rate with or without warfarin in patients with Hancock mitral valve prostheses. 2. Warfarin significantly increases major bleeding complications in this clinical setting.

*By invitation

31. Year of Operation as a Risk Factor in the Late Results of Valve Replacement

ALBERT STARR, GARY L. GRUNKEMEIER*,

JOSEPH F. TEPLY* and QUENTIN MacMANUS*,

Portland, Oregon

The actuarial thromboembolic rates of aortic and mitral noncloth-covered caged-ball valves used during the second decade of cardiac valve replacement are significantly lower than for the same prostheses implanted during the first decade, as shown in the following table:

Model	Mitral 6120		Aortic 1200/60
Implant years	1965-72	1973-79	1965-72	1973-79
Number of patients	83	67	133	216
Maximum/mean follow-up (years)	14/8.4	5/1.8	13/8.8	6/1.9
Five year embolus-free rate ( ± SE)	69(± 6)%	96(± 4)%	78(± 4)%	93(± 2)%
P-value	p=.017		p=.004

Five year embolus-free rates for the composite-strut caged-ball, Bjork-Shiley tilting disc and porcine xenograft valves all fall in the range of from 88% to 91% for the mitral position and from 91% to 94% for the aortic. Thus the standard noncloth-covered prosthesis, used during the current era, has a thromboembolic risk as low as that reported with other concurrently utilized valve substitutes. This striking reduction in thrombogenicity demonstrates that the time frame of implantation must be considered when evaluating the results of cardiac valve replacement.

*By invitation

32. Open Mitral Commissurotomy-A Modern Evaluation

WILLIAM L. HALSETH*, DONALD P. ELLIOTT*,

E. LANCE WALKER* and ELEANOR A. SMITH*,

Denver, Colorado

Sponsored by: Ben Eiseman, Denver, Colorado

Familiarity with replacement of the mitral valve (MVR) with prosthetic and tissue valves has dimmed awareness of the usefulness of open mitral commissurotomy (OMC). This is a review of a 10-year experience ending in December 1978 of 259 consecutive patients operated upon with a clinical diagnosis of mitral stenosis. MVR was necessary in 62 patients (24%), primarily because of severe deformity of valvular and subvalvular structures. No closed commissuortomies were performed-an operation now considered passe.

Of the 197 OMC, 12 had additional cardiac procedures. Of the 3 patients who died (1.52%), two were operated upon as emergencies because of rapidly progressive cardiac failure.

Followup was obtained on 191/197 (91%)of the OMC patients. Late mortality was 9%(18 patients) of whom 14 were cardiac related. 76% of patients (146) had at least one category New York Heart Association Class improvement following OMC. Fourteen (7%) of the 191 OMC patients had subsequent MVR at times varying from 2-92 months (mean 41.6 months). Ten year survival for the 191 OMC patients was 81%.

Summary and Conclusions:

(1) Open mitral commissurotomy (OMC) was performed in 197 of 259 patients with the clinical diagnosis of mitral stenosis. Of these, 81% survived 10 years, and 14 (7%) subsequently required valve replacement. (2) OMC is the operation of first choice for patients with the clinical diagnosis of mitral stenosis. The main indication for proceeding to MVR at the first operation is severe valve deformity, which prevents successful commissurotomy.

(3) There will be no need for valve replacement in 93% of patients following OMC, 76% of whom will experience measurable and lasting functional benefit from open commissurotomy alone.

(4) This clinical experience emphasizes that open commissurotomy rather than valve replacement is the best initial treatment for most patients with mitral stenosis.

*By invitation

33. Is Tricuspid Valve Repair Necessary?

CARLOS M. G. DURAN, JOSEL. POMAR*,

THIERRY COLMAN*, ALVARO FIGUEROA*

JOSE M. REVUELTA*, and JOSE L. UBAGO*, Santander, Spain

In an attempt to clarify the still confusing indications for tricuspid surgery, a group of 150 patients (pts) has been studied. Selection was made according to the following criteria 1) All had preoperative significant tricuspid disease and 2) pre and postoperative right and left catheterization including biventriculography was available. It has been previously proven in our laboratory, by correlating the hemodynamic and surgical findings, that our technique of right ventriculography is highly reliable in the diagnosis of tricuspid insufficiency (TI) and whether it is an organic or functional lesion. 123 (82%)pts were recatheterized as part of a routine follow up study and 27 (18%) for clinical deterioration. 89 had an associated mitral and 61 mitral and aortic disease. After left side repair, 119 pts underwent tricuspid surgery (group A). In 31 pts the tricuspid valve was not repaired (group B).

In group A, 64 (54%) had one or more tricuspid commissures ("organic") and the remainder 55 were classified as "functional". Tricuspid valve repair includes 73 ring annuloplasties, 42 commissurotomies and annulo-plasties and 4 commissurotomies. Postoperative gradients averaging 2.8 mm Hg (± 1.6) were found in 31% of the pts with functional and in 44% of those with organic TI. Those pts with tricuspid gradients averaged a cardiac index (CI) of 3.2 L/min/m² and those without gradients 2.8 (p<0.01). Significant residual TI was found in 10 functional and 26 organic valves. Clinically, 85% were in Functional Class I despite their small residual gradients and TI. Further analysis of the hemodynamic data of this group of pts showed that the postoperative CI was directly related to the adequacy of the left side repair and to the residual pulmonary resistances independently of whether residual TI was present or not. These facts seem to imply that the associated tricuspid repair is at best unnecessary. However, when measuring right ventricular volumes it was observed a postoperative increase of 9.3% in those pts with persistent TI and a 37% decrease in those with tricuspid competence.

In group B, all 14 pts with organic disease showed residual TI. 8 pts out of the 17 with functional TI had no regurgitation concomitant with a significant decrease in pulmonary resistances. In the remainder 9 pts the TI persisted.

Based on these data that emphasize the need for a correct left side repair, our present attitude towards tricuspid repair is: 1) Functional TI can only be ignored in pts with predictable significant reduction in pulmonary resistances, 2) Organic disease must be repaired.

INTERMISSION - 45 MINUTES

VISIT EXHIBITS

*By invitation

34. Sternal Wound Complication Management And Results

CYRUS SERRY*, PHYLLIS C. BLECK*, HUSHANG JAVID,

JAMES A. HUNTER, MARSHALL D. GOLDIN*,

GIACOMO A. DELARIA*, and HASSAN NAJAFI,

Chicago, Illinois

Records of 4,125 patients who underwent sternotomy for cardiac surgery were reviewed to determine the incidence of sternal wound complications (1.8%), including wound drainage, skin separation, unstable sternum, and sternal dehiscence with or without infection. Septicemia and mediastinal abcesses were found in all 19 patients who expired.

Based on findings, local treatment (incision and drainage of skin and subcutaneous tissue with frequent change of dressing or irrigation (Method A) is recommended for (1) serosanguineous drainage, (2) patients with stable sternums and superficial infection, but no systemic reaction. However, surgical debridement and closure followed by mediastinal irrigation via drainage tubes with 0.5% povidone iodine solution (Method B) is recommended for (1) draining, unstable sternums, (2) wound infections involving the retrosternal space and (3) wound infections causing systemic reactions unresponsive to Method A.

It was found that none of the eight patients in the latter group died when managed by Method B, and only one developed recurrent infection. In contrast, from 28 patients of the latter group not treated with Method B, 11 died of infection related causes and 13 returned with recurrent infection.

35. Fulminating Non-Cardiogenic Pulmonary Edema-A Newly Recognized Hazard During Cardiac Operations

ALFRED T. CULLIFORD,* STEPHEN THOMAS* and

FRANK C. SPENCER, New York, New York

In the past 24 months, four patients have been successfully treated for fulminating pulmonary edema following coronary bypass operations. In each patient, the fulminating edema began within 30-60 minutes after infusion of blood products, usually fresh-frozen plasma, suggesting an allergic mechanism. Copious frothy fluid poured from the endotracheal tube, with hypoxia and respiratory acidosis. Left atrial pressure was only 5.0-10 mm., excluding left ventricular failure; and the electrocardiogram and CPK isoenzymes excluded myocardial infarction. The albumin content of the fluid was significantly elevated, suggesting an exudate rather than a transudate.

Three postoperative patients were promptly reoperated upon, and in one of these, the sternum could not be closed for 72 hours because stiff, distended lungs compressed the heart. Treatment consisted of intravenous prednisolone 2.0-3.0 gms., positive pressure ventilation, diuretics and albumin (in the last two patients).

It was dramatic that within 72 hours this near-lethal condition subsided completely, with no further cardiac or pulmonary complications.

The allergic mechanism is still unclear, but the effectiveness of immediate treatment, as described, prompts this early report. Undoubtedly, patients have died in past years with the erroneous diagnosis of fulminating cardiac failure or "pump lung." This rare condition is readily recognized when massive pulmonary edema develops without signs of left ventricular failure-an occurrence which is easily discerned in a properly monitored patient.

*By invitation

36. Immediate Coronary Artery Bypass for Acute Evolving Myocardial Infarction (AEMI)

RALPH BERG, JR., SAMUEL L. SELINGER*,

RONALD P. GRUNWALD* and

WILLIAM P. O'GRADY*, Spokane, Washington

212 consecutive patients (March 1971-April 1979) all had infarction syndrome, electrocardiographic, coronary angiographic, ventriculographic and retrospective enzyme changes consistent with AEMI. In 212 AEMI patients, means age 55.8 years (R = 28-79), onset of CAB surgery was less than six hours, yielding three surgical deaths for a 1.7% mortality. Conversely, medical therapy in 200 consecutive AEMI patients, mean age 53.2 years (R = 40-65) resulted in 23 deaths giving in-hospital mortality (IHM) of 11.5% plus first year mortality of 14%. Pre and post-operative enzyme and EKG values, age, sex and coronary artery involvement are presented. Angiograms done an average of 7.9 months post-op of 97 grafts to the infarcting areas revealed 95 patent (92%). Ejection fractions were normal, unchanged or improved in 86%. Two small ventricular aneurysms were noted. Follow-up of 200 patients revealed three deaths within the first year (1.8%). 26 patients had mild angina. Quality of life was the same or improved in 196. In summary, AEMI patients must have CAB early in the syndrome. Surgical results are far superior to medical therapy: Medical IHM (11.5%) + first year mortality (14%) = 25.5%. Surgical IHM (1.7%) + first year mortality (1.8%) = 3.5%.

*By invitation

37. Multivariate Discriminant Analysis of Clinical and Angiographic Predictor of Coronary Surgery Mortality

J. WARD KENNEDY*, GEORGE KAISER, LLOYD FISHER*,

WILLIAM O. MYERS, GERARD MUDD* and

THOMAS J. RYAN*, Seattle, Washington; St. Louis, Missouri

CASS is a large multi-institutional study of the medical and surgical treatment of coronary artery disease. Fifteen cooperating institutes have operated on 7,122 patients from August, 1975 through December, 1978. The in hospital operative mortality (OM) was 2.89%. In an effort to better understand the clinical and angiographic characteristics predictive of OM, we have done a multivariate discriminant analysis of variables associated with OM.

Numersou clinical and angiographic variables were selected from the CASS data file and evaluated in a univariate manner for their relationship ot OM. Twenty-two of these variables were then selected for multivariate discriminant analysis. Clinical variables of most value were age, sex, heart size, and symptoms, signs, or treatment for congestive heart failure. Angiographic variables of importance included EF, left ventricular wall motion abnormalties, and the severity of left main coronary disease.

Multivariate discriminant analysis resulted in the selection of eight variables that contained the most predictive information as listed below.

Variable	F Statistic
Congestive Heart Failure Score	73.3
Age	37.5
Left Main Coronary Stenosis ≥ 90%	24.8
Left Ventricular Wall Motion Score	22.9
Female Sex	21.4
Heart Enlargement by Chest X-Ray	17.9
Pulmonary Rales	12.0
Left Ventricular Dilitation by Angiography	9.4

The strong association of OM with advanced age, female sex and variables associated with left ventricular dysfunction is clear. The risk of OM for an individual patient may be estimated with the use of these eight clinical and angiographic characteristics.

*By invitation

38. Doppler Velocity Measurements of Coronary Flow: A Noninvasive Intraoperative Guide to Hemodynamically Significant Lesions

CREIGHTON B. WRIGHT, DONALD B. DOTY,

MELVIN L. MARCUS*, and CHARLES L. EASTHAM*,

Iowa City, Iowa

Although angiography is in general a reliable guide to the coronary vessels which should by bypassed in individuals with angina pectoris, some lesions escape accurate preoperative assessment. This leaves the surgeon with the dilemma of whether to bypass all visible vessels or to risk not bypassing significant disease.

Utilizing the miniaturized pulsed single crystal Doppler velocity probe and technique developed at the University of Iowa Hospitals and Clinics to measure phasic coronary velocity, the hemodynamic significance of experimental and clinical coronary stenoses can be assessed readily. No dissection of the vessels is required. The probe is held in place over the coronary vessel with a small suction cup. The velocity traces are linearly related to electromagnetic blood flow (R = 0.98). Reactive hyperemia curves obtained with 20 second occlusions demonstrate reproducibly the vessels with hemo-dynamically significant stenoses.

This technique has been used in over 50 patients with a variety of anatomic lesions. In 6 recent patients in which the degree of stenosis of a vessel was disputed or thought to be less than 50%, the use of the Doppler probe and a 20 second reactive hyperemia curve allowed accurate and immediate assessment of the need for bypass.

It is the purpose of this presentation to describe this new unique technique, its benefits, limitations, and overall accuracy.

Adjournment

1:00 P.M. Cardiothoracic Residents' Luncheon

Continental Ballrooms 8 & 9

*By invitation