HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Arom, K. V. Articles by Grover, F. L. Search for Related Content PubMed PubMed Citation Articles by Arom, K. V. Articles by Grover, F. L. Related Collections History

Ann Thorac Surg 2003;76:S17-S46
© 2003 The Society of Thoracic Surgeons

---

Supplement

Adult cardiac surgery during the first 50 years of the Southern Thoracic Surgical Association

^a Bangkok Heart Institute, Bangkok, Thailand
^b University of Colorado Health Sciences Center and Denver VA Medical Center, Denver, Colorado, USA

^* Address reprint requests to Dr Grover, University of Colorado Health Sciences Center, Division of CT Surgery (C-310), 4200 East Ninth Ave, Denver, CO 80262, USA
e-mail: frederick.grover{at}med.va.gov

Presented at the 50th Anniversary of the Southern Thoracic Surgical Association, Bonita Springs, FL, Nov 14, 2003.

Abstract

This is a review of some of the outstanding adult cardiac surgical papers presented during the first 50 years of the Southern Thoracic Surgical Association to commemorate the 50th Anniversary meeting. Since the founding of the Southern Thoracic Surgical Association in 1954 there have been a total 512 adult cardiac surgery and great vessels papers presented, 138 from 1954 to 1970, 157 from 1971 to 1986, and 217 from 1987 to 2002. Since 1964 most of these papers have been published in The Annals of Thoracic Surgery. Forty-three papers were reviewed, the first having been presented in 1956 and the last in 2002. Not all of the papers could be located but the vast majority were retrievable and reviewed. This paper presents a broad spectrum of adult cardiac surgery beginning with early pump technology, early myocardial revascularization including the Vineberg procedure followed by coronary bypass as we know it today, valvular surgery, including several papers on aortic valve homograft and autograft procedures, ventricular aneurysms, aortic aneurysms and aortic dissections. Evaluations of various valve prostheses, cardiac transplantation, mitral valve reconstruction, quality assurance in cardiac surgery, neurobehavioral outcome after cardiac surgery, endocarditis and off-pump coronary bypass surgery were reviewed. We hope that this article is representative of the broad spectrum of issues that have characterized the specialty of cardiothoracic surgery over the second half of the 20th century and the beginning of the 21st century.

During the period 1950 through 1952 many discussions were held concerning the formation of a regional association for thoracic surgery. The letter over the signatures of Dr James D. Murphy, Dr Dewitt C. Daughtry, and Dr Hawley H. Seiler was sent to 66 thoracic surgeons in southern states suggesting a meeting to discuss the formation of a regional society whose membership would be limited to qualified thoracic and cardiovascular surgeons in the southern states [1].

Year nineteen hundred and fifty-four was a very important year for thoracic surgery and cardiovascular surgeons. The closure of simple atrial septal defect had been going well after John F. Lewis accomplished the first successful repair on September 12, 1952, under hypothermia with the heart beating. That was followed by the successful closure of a simple atrial septal defect by Dr John Gibbon on May 6, 1953, using the heart-lung machine that he developed [2–4]. Doctor C. Walton Lillehei first successfully closed a ventricular septal defect using cross circulation in April 1954. All of those accomplishments led to the successful modern era of cardiac surgery.

Year nineteen hundred and fifty-four was also a milestone for the Southern Thoracic Surgery Association. The first annual meeting of this association was held at the Hollywood Beach Hotel in Hollywood, Florida, on December 5 and 6, 1954, and this was an outstanding success. Fifty-four surgeons were present and there were 17 scientific presentations, all of excellent quality. Two papers were concerned with cardiac and great vessel surgery [1]. Since then there has been a tremendous growth of cardiac and great vessel abstracts submitted and accepted to be presented at each annual meeting [5]. We were charged by the 50th anniversary organizing committee to review the interesting and outstanding papers contributed to our cardiac surgery field presented at our annual meetings during the past 50 years.

Material and methods

The Southern Thoracic Surgical Association program books from 1954 to 2002 were successfully retrieved, except for one in 1964. This program book after extensive searches by several parties is still not available.

Since 1957, the fourth annual meeting of the association, the printed program was no longer a small leaflet but a 24-page booklet containing abstracts of the scientific papers. During the spring of 1964 the matter of publishing our annual meeting papers was resolved in a most satisfactory manner. The Annals of Thoracic Surgery, the journal sponsored by the newly formed Society of Thoracic Surgeons, was designated our official journal.

At the meeting held in Freeport, Grand Bahamas Island, in 1965 our membership exceeded 250 for the first time [1]. From 1965 on the Southern Thoracic Surgical Association continued to mature and grow in stature among the surgical societies in this country. The scientific presentations continued at a high level of excellence, attracting many young surgeons to join [5].

To the best of our knowledge, there were 512 cardiac surgery and great vessel papers presented and available to review. We have divided these papers into three eras according to the time frame: the early years (1954 to 1970), 138 papers; the middle years (1971 to 1986), 157 papers; and the recent years (1987 to 2002), 217 papers.

The subjects were divided into seven categories: coronary, myocardium, and revascularization; valves; aorta and great vessels; extracorporeal circulation; trauma; transplantation; and other. The numbers of papers in each category and the totals are listed in Table 1 for the early years; in Table 2 for the middle years, and in Table 3 for the recent years.

During the early years three institutions were tied for first place with five papers each. Four institutions were tied for second place with four papers each and three institutions were tied for third place with three papers each (Table 4). During the middle years, first place went to The John Hopkins Hospital and School of Medicine in Baltimore with 12 papers. Second place went to Baylor College of Medicine in Dallas and Vanderbilt University, each with nine papers. Emory University won the third place alone with eight papers (Table 5). For the recent years, there was no tie. Thirteen papers came from Johns Hopkins Medical Institute, 12 papers from the University of Virginia Health Science Center (Charlottesville), and 11 papers from the University of Texas Medical School in Houston (Table 6). The states in each era most often contributing and presenting papers are listed in Tables 7, 8, and 9. The top three first authors/presenters in each era are listed in Tables 10, 11, and 12.

Sealy, Brown, Young, Stephen, Harris, and Merritt [6] reported on hypothermia low-flow extracorporeal circulation and controlled cardiac arrest for open heart surgery. These colleagues presented the Duke experience with hypothermic extracorporeal circulation and controlled cardioplegia for open heart surgery at the 1956 meeting of the Association. These authors described a technique where they used moderate hypothermia combined with low-flow perfusion to extend the period of time that they could repair cardiac defects. They did this because at that particular time high flows were hard to obtain and also partly because of the complexity of the equipment and also the deleterious effects of the perfusion systems on blood. In addition these authors described using what we would now call cardioplegia, a solution of potassium, magnesium, and prostigmin injected into the aorta to create standstill. The authors report six cases where they packed their patients in ice after the induction of anesthesia, cooled to a rectal temperature of 34°C, then removed the patient from the ice and began the operation. Usually the patient's core temperature drifted down to 30 to 31°C. They cannulated the inferior vena cava through the saphenous vein; the superior vena cava through the right subclavian vein and arterial cannulation was through the left subclavian artery. They had a special system of plastic bags which served as oxygenators, putting oxygen in a bag along with a unit of blood, shaking them, and then sequentially perfusing them into the patients. (Fig 1). This required a reservoir of 2 to 5 L. Flows were in the range of 20 to 40 mL per kilogram of body weight. Pump time was up to 21 minutes. This was one of the earliest papers describing the combination of hypothermia and a perfusion system and was advocated for extending the time to perform intra cardiac procedures.

View larger version (43K): [in this window] [in a new window]   Fig 1. This diagram demonstrates the time required to oxygenate and perfuse a patient with blood at 500 ml per minute with the author's method of extracorporeal circulation. (Reprinted with permission from Sealy et al, Surg Gynecol Obstet 1957;104:441–50.)

Yeh, Anabtawi, Cornett, White, Stern, and Ellison [8] reported at the 1966 annual meeting of the Association in Ashville, North Carolina, their experience with bacterial endocarditis after open heart surgery. The authors reviewed 400 open heart procedures operation from November 1957 to August of 1966, 220 of which were congenital and 180 were acquired cardiac disease. Blood cultures were obtained routinely from pump equipment before and after bypass and from the patient after surgery daily for 3 days or longer if indicated. At that time prophylactic antibiotics consisting of procaine penicillin and streptomycin were started on the day before and were continued for 5 days after surgery. During the last year of the study this was modified to include oxacillin 4 g daily for 7 days.

Of interest is the fact that a history of bacterial endocarditis or a positive culture from the pump equipment before bypass did not predispose patients for postoperative infection. Nineteen patients who had a positive culture from the pump oxygenator after bypass developed clinical manifestations of endocarditis. If direct repair of a lesion was utilitized the incidence of endocarditis was only 1.3%. However if a protheses was used other than a ball-valve it was 3.5% and with a ball-valve protheses the incidence was 9.9%. Half of the infections were due to Staphylococcus species. Of a total of 18 cases, nine were clinically cured and nine patients died. The authors noted that although the mortality rate for an infected ball-valve prothesis was high, some of these patients could be salvaged with proper antibiotic therapy and occasional removal and replacement of the prosthesis. This paper emphasized the seriousness of postoperative endocarditis and described prophylactic and therapeutic antibiotic regimen for that period.

Cooley, Bloodwell, and Hallman [9] reported on their clinical experience with cardiac transplantation at the 1968 meeting of the Association. They described four patients who were in end-stage heart failure who underwent orthotopic cardiac transplantation. At the time of their report, two patients were alive and well 6 weeks and 3 and a half weeks postoperatively. Donors had end-stage irreversible brain injury, both blunt and penetrating. One had a spontaneous intracranial hemorrhage. Immunosuppressive agents that were used included corticosteroids, azathioprine, and antilymphocyte globulin. The authors note that they had no direct evidence of any serious rejection and concluded that their experience suggested that human cardiac transplantation was feasible and deserved further clinical trials.

These authors used the classic approach of Shumway, utilizing the right atrial side-to-side anastomosis as opposed to the bicaval technique (Figs 2, 3, and 4). Of interest is that in that era the donor was heparinized, the heart was removed in a similar fashion to that done today and placed in a sterile basin without the use of any special preservation techniques, and then taken to an adjoining operating room for implantation. The donor heart ischemic time was approximately 45 minutes in these cases. The transplants performed were based on ABO blood group compatibility and in one case tentative histocompatibility was established before transplantation. The use of antilymphocyte globulin was the first employment of that agent reported in human heart transplantation. This series is one of the first published on human heart transplantation and marked a new era for cardiothoracic surgery.

View larger version (59K): [in this window] [in a new window]   Fig 2. Demonstration of the cannulation technique utilized for heart transplantation by Cooley and colleague's bicaval and femoral artery cannulation with the insert demonstrating the excision of the recipient's diseased heart leaving maximum aorta and main pulmonary artery and left and right atrial cuffs. (Reprinted with permission from Cooley et al, JAMA 1968;205:67–74. Copyright American Medical Association 2003. All rights reserved.)

View larger version (87K): [in this window] [in a new window]   Fig 3. Preparation of the donor heart with adequate aortic and pulmonary artery segments, ligation of the superior vena cava just superior to the caval atrial junction, and incision of the right atrium from the inferior vena caval orifice toward the appendage, care being taken to avoid the sinus node. The superior and inferior pulmonary veins are connected on each side and then a transverse incision is made to create a common atrium. (Reprinted with permission from Cooley et al, JAMA 1968;205:67–74. Copyright American Medical Association 2003. All rights reserved.)

View larger version (62K): [in this window] [in a new window]   Fig 4. The suturing of the atrial component of the procedure with the left atrium being sutured first, then the septum and the right atrium. The pulmonary artery and aorta are anastomosed end-to-end. (Reprinted with permission from Cooley et al, JAMA 1968;205:67–74. Copyright American Medical Association 2003. All rights reserved.)

In this series of 535 patients who underwent coronary arteriography, there were three late deaths not attributed to the procedures and 21 serious complications. It is of interest that the authors state, "true significance for each individual patient of what is demonstrated on the coronary arteriogram is still very much a matter of conjecture. Arteriographic findings cannot stand alone but must be correlated with all other available information in each individual instance." This presentation indicates the active involvement of cardiothoracic surgeons in performing diagnostic cardiac catheterization at that time, involvement that was subsequently relinquished.

Diethrich, Liddicoat, Alessi, Kinard, and De Bakey [11] reported on serum enzyme and electrocardiographic changes immediately after myocardial revascularization at the 1968 meeting in Dallas, Texas. The authors reviewed their experience with 40 patients who underwent double internal mammary artery implantation using the Vineberg internal mammary artery-myocardial tunnel technique. All had severe angina, the majority having angina at rest. Twenty-five patients had a history of at least one myocardial infarction. There was one postoperative death on postoperative day 11 from a massive myocardial infarction. The authors were particularly interested in evaluating the incidence of postoperative myocardial infarction and better defining it. They acknowledged that they had difficulty with defining myocardial infarction after this procedure since a tunnel was made in the myocardium, which in itself creates injury to the myocardium. The patients were divided into three groups based on their preoperative electrocardiogram. Nine patients were in group I with normal preoperative electrocardiograms, eight in group II with abnormal electrocardiograms, and 23 in group III with definite evidence of a prior myocardial infarction. In group I, two of nine patients converted to an electrocardiogram demonstrating a myocardial infarction; in group II, three of eight patients demonstrated a myocardial infarction; and in group III, three of 25 had definite changes, one probable and two suggestive.

The authors then examined enzymatic findings and found that only three patients had entirely normal postoperative enzymes. They then attempted to correlate the electrocardiograms and enzymes and found that of the 11 patients who had an ECG suggestive of a myocardial infarction, eight had elevated enzymes and one of these was the patient who died. This is one of the earliest papers attempting to assess the incidence of perioperative myocardial infarction associated with coronary surgery and illustrates the difficulty of doing so, which still remains 35 years later.

Johnson, Flemma, and Lepley [12] reported on direct coronary surgery utilizing multiple-vein bypass grafts at the 1969 meeting of the Southern Thoracic Surgical Association in Washington, DC. This report describes experience with 90 patients who underwent double and triple coronary bypass with vein grafts for an 11-month period. Patients' ages varied from 28 to 66 years; 15 had only two-vessel disease, whereas 75 had greater than 75% occlusion of all three major arteries. The contraindications to operation were current smoking, overt cardiac failure in patients more than 70 years of age, and other organ deterioration. Additional procedures such as aneurysmectomy or multiple arterial implants were performed simultaneously in 40% of patients. Veins were anastomosed in an end-to-side fashion to the coronary arteries and attached directly to the aorta. The authors utilized electrical fibrillation with intermittent 15-minute intervals of anoxic, hypothermic arrest. More than 80% of their vein grafts were restudied. In one was there a mild anastomotic stenosis and more than 90% were patent. The authors reported an overall mortality rate of 22% and noted that 40% of the patients would have been candidates for transplantation at other centers. Double-vein bypasses were carried out for two-vessel disease with a mortality of only 7%. The authors point out that direct coronary bypass immediately increases myocardial blood flow to multiple areas of ischemic myocardium as opposed to the previously utilized Vineberg procedure, which took weeks to months to develop collaterals. Their extension of the coronary bypass operation to triple-vein grafts was first reported in this paper and this established an important milestone. Figure 5 demonstrates a saphenous vein bypass graft to a distal right coronary artery and one to a left anterior descending coronary.

View larger version (185K): [in this window] [in a new window]   Fig 5. An illustration of two injections of saphenous vein grafts from a cardiac catheterization. On the left side of the figure a right saphenous vein graft into the distal right coronary artery is demonstrated, and on the right side of the figure a saphenous vein graft into the left anterior descending (lad) coronary artery is demonstrated. (Reprinted with permission from Johnson et al, Ann Thorac Surg 1970;9:436–44.)

Robicsek, Tam, Daugherty, and Mullen [14] described their studies involving myocardial protection during open heart surgery at the 1969 of the Association in Washington, DC. The authors stated that at that time direct cannulation and selective perfusion of the coronary arteries had become a preferred method of myocardial protection. Although they acknowledged that this technique had well known benefits, they were concerned about a number of disadvantages including the deprivation of a dry and quiet operative field and the additional time required to properly position cannulas. They noted that the cannulas are frequently in the way of the operating surgeon, that the coronary perfusion created an excessive amount of suctioned blood which led to hemolysis and also the likelihood of introduction of bubbles, emboli or particulate matter into the coronary circulation. In addition they noted that on the left side the catheter could be directing flow into the circumflex system, excluding the anterior descending coronary artery or vice versa. Another complication they cited was arterial dissection from the cannulas or damage to the intime at the coronary orifices.

They reported that during the past 7 years they had performed cardiac procedures using high-flow, normothermic total-body perfusion with complete interruption of the coronary circulation, and external (slush) cooling of the heart. During that period they operated on 342 patients with acquired and 121 patients with congenital heart disease. In three patients the heart could not be resuscitated and eight patients died in the operating room (2.5%). They attributed these deaths to primary myocardial insufficiency. They reported a total of 11% operative mortality for their entire group, 11.1% for patients with acquired heart disease, and 13.1% for patients undergoing heart valve replacement.

These findings led the authors to perform a group of animal studies comparing direct coronary perfusions to cross clamping and topical hypothermia. The authors noted that the animals that had direct perfusion of the coronary arteries had an additional 5.5 minutes of perfusion required for the coronary cannulation. They noted that hemodynamic impairment occurs even if the coronary arteries are cannulated and perfused but appeared to be less pronounced and of shorter duration if the myocardium was protected by topical hypothermia rather than by coronary perfusion during a 45-minute period of aortic cross-clamping. Finally the authors suggested that there be reconsideration regarding artificial perfusion of the coronary arteries as the preferred method of myocardial protection, and that thought be given to ischemic arrest and topical hypothermia with saline slush.

Gonzalez-Lavin, Al-Janabi, and Ross [15] presented their long-term results after aortic valve replacement with preserved aortic homografts at the 1971 Association meeting in Tampa, Florida. The authors reported on 259 patients who had undergone aortic valve replacement with preserved homografts with a total mortality of 16.6%, which declined to 4.8% for the last two years (1968 and 1969). The series included patients operated on from early 1964 to late 1969 at the National Heart Hospital in London. Seventy-one percent of patients still had their original homograft valve at the time of the writing of the paper. Valve failure occurred in only 46 of 259 patients and it was due to cusp rupture and calcification in 28 of those patients. Late mortality related to valve failure was 5.4%. At the time of the report the authors had switched to sterilizing homografts with antibiotics in an attempt to avoid degenerative changes by implanting a fresh and viable valve. 57.5% were operated for aortic stenosis and 42.5% for aortic regurgitation. The authors noted that since June 1967 the aortic homografts were not used in most patients less than 45 years of age with isolated valve disease but rather what we now know as the Ross procedure was used in that group. Of interest is the fact that mechanical prostheses were used mainly in older persons in the seventh and eighth decades of life. The great majority of valves utilized in this study were sterilized by ethylene oxide and preserved by freeze drying. The valves were then put in the anatomic position with two suture lines, one at the annulus and one above the cusp to attach the homograft aortic wall to the wall of the host sinuses of Valsalva. A good functional result was obtained in 95% of 212 patients discharged from the hospital.

Patients were followed up for 1.5 to 7.5 years (Fig 6). Thirteen and a half percent of patients died during the follow-up period and the late mortality as mentioned for valve failure was 5.4%. Of the 151 patients who at the time of the analysis retained their original grafts, only 4.6% had moderate aortic regurgitation. Valve failure occurred in 17.8% of the 259 patients. The failure rate was 25.5% on those operated on from 1964 to 1966 but from 1967 to 1969 valve failure averaged only 8.4%. Thirteen patients required reoperation for removal of a calcified homograft. Figure 6 demonstrates the clinical status of 191 patients discharged from the hospital with a preserved homograft aortic valve and followed up for 1.5 to 7.5 years. The late mortality from nonvalve unrelated causes was excluded in this drawing. Figure 7 demonstrates the clinical status of patients followed from 4.5 to 7.5 years postoperatively. Figure 8 demonstrates the survival curves for patients discharged from the hospital and maintaining a good functional result for the year of operation. The authors acknowledge that the incidence of valve failure, although low was still an important problem and was usually due to technical error or degenerative changes. They also noted that the method of valve preparation and preservation is extremely important. They noted that "perhaps a fresh and viable homograft valve may become a permanent valve substitute." This was a seminal paper demonstrating long-term results with aortic homografts by the group that pioneered this effort.

View larger version (46K): [in this window] [in a new window]   Fig 6. The current clinical status of 191 patients discharged from the hospital with a preserved homograft aortic valve who were followed up 1.5 to 7.5 years. Twenty-one patients who died of causes unrelated to homograft malfunction were excluded. (Reprinted with permission from Gonzales-Lavin et al, Ann Thorac Surg 1972;13:594–606.)

View larger version (43K): [in this window] [in a new window]   Fig 7. The clinical status of 102 patients who were discharged from the hospital with a preserved homograft valve who were followed up from 4.5 to 7.5 years. Twelve patients who died of causes unrelated to homograft malfunction were excluded. Note that at the latter end of the 7.5 year follow-up, there is a somewhat greater incidence of homograft failure (33% versus 20%). (Reprinted with permission from Gonzales-Lavin et al, Ann Thorac Surg 1972;13:594–606.)

View larger version (11K): [in this window] [in a new window]   Fig 8. Long-term results with preserved aortic homografts. The survival curve for patients discharged from the hospital and maintaining a good functional result. There is a trend toward improved results with those operated on during the later periods. (Solid line = year of operation 1967; dashed line = year of operation 1966; dotted line = year of operation 1965). (Reprinted with permission from Gonzales-Lavin et al, Ann Thorac Surg 1972;13:594–606.)

View larger version (50K): [in this window] [in a new window]   Fig 9. Resection of a left ventricular aneurysm is demonstrated and the lower right figure shows the advancement of the lateral myocardial wall down into the ventricular septum, to help cardiac function. (Reprinted with permission from Stoney et al, Ann Thorac Surg 1973;15:394–404)

View larger version (25K): [in this window] [in a new window]   Fig 10. Reduction in left ventricular end diastolic pressure after aneurysm resection and repair in 20 patients. (LVEDP = left ventricular end-diastolic pressure; POST-OP = postoperative; PRE-OP = preoperative; * = patient who died.) (Reprinted with permission from Stoney et al, Ann Thorac Surg 1973;15:394–404)

Trinkle, Marcos, Grover, and Cuello [17] at the annual meeting in 1973 in Louisville, Kentucky, reported on management of the wounded heart. The authors reviewed 45 patients who were treated for penetrating trauma over a 5-year period beginning in July of 1968. Figure 11 illustrates the location of the injuries. There were five early deaths (11%), three of which were due to uncontrolled bleeding from associated injuries. The combined early and late mortality was 15.5%. The authors attribute part of the low mortality in this series to natural selection (only patients arriving with signs of life were included in the study) and the fact that 30 of the patients had stab wounds and only 13 had gunshot wounds. Thirty-six patients presented in cardiac tamponade and nine in hemorrhagic shock. The authors note that patients with the most severe and irreparable injuries were usually dead on arrival. Of the 45 patients reaching the hospital alive, every one of the heart and intrapericardial vessels was successfully repaired without the use of cardiopulmonary bypass. None of the deaths were caused by the inability to control the intrapericardial wound.

View larger version (37K): [in this window] [in a new window]   Fig 11. The site of injury in 45 patients with penetrating wounds to the heart and great vessels is demonstrated. (Ao = aorta; IVC = inferior vena cava; LV = left ventricle; PA = pulmonary artery; PV = pulmonary vein; RA = right atrium; RV = right ventricle; SVC = superior vena cava.) (Reprinted with permission from Trinkle et al, Ann Thorac Surg 1974;17:230–6.)

View larger version (53K): [in this window] [in a new window]   Fig 12. Various technical maneuvers useful in the repair of penetrating injuries of the heart and great vessels are shown including (A) digital pressure, (B) using crossing sutures to initially control hemorrhage, (C) partial occlusion clamps for atrial and great vessel injuries, (D) electrical fibrillation to achieve cardiac standstill, (E) inflow occlusion, and (F) pharmacologic hypotension. (Reprinted with permission from Trinkle et al, Ann Thorac Surg 1974;17:230–6)

The authors reported on 88 patients with arteriographic findings of significant coronary artery disease and a clinical picture of unstable angina pectoris. The decision to operate or manage by nonoperative therapy was made by the attending physician. Fifty-three patients underwent coronary artery bypass grafting (CABG) with saphenous veins and 35 were treated medically. There was a 5.7% 30-day mortality in both groups. However there was a marked difference in improvement of symptoms after the first 30 days, with most of the surgically managed patients either asymptomatic or greatly improved after coronary bypass whereas two thirds of the medically treated patients had persistent severe angina pectoris. In addition there were two late deaths in the medical group and none in the surgical group. The authors concluded that coronary bypass could be performed safely during the unstable phase of coronary artery disease and that relief of angina could be anticipated, as compared with those treated medically where there was an appreciable risk of death and the majority of patients continued to experience angina. This article therefore is one of the first demonstrating the safety and efficacy of performing coronary bypass on patients with unstable angina.

Parker, Neville, Hanson, Mohiudan, and Webb [19] reported on their experience in the management of acute aortic dissection in 1974 at the 21st annual meeting of the Southern Thoracic Surgical Association in Williamsburg, Virginia.

The authors observed an excessive operative mortality in patients with acute dissection who were operated upon within 24 hours of admission. They therefore began utilizing intensive medical management to delay surgical intervention to at least three weeks or longer in type III dissections unless other indications arose. Clinical management consisted of control of hypertension with close invasive monitoring with indications for operation being occlusion of any major branches of the aorta or failure to control pain or hypertension. The authors noted a mortality in the combined medical-surgical group of 19% as compared with 75% when operating acutely. After that experience they developed a delayed operative intervention policy and had no early deaths. This included two type I, seven type II, and four type III dissections using the DeBakey classification. They utilized aortic valve resuspension when possible and necessary in type I and II aneurysms (Fig 13) and utilized a graft interposition in the ascending aorta in this group of patients (Fig 14).

View larger version (79K): [in this window] [in a new window]   Fig 13. The technique for resuspension of the aortic valve after acute dissection of the ascending aorta is demonstrated with each commissure being anchored by a Teflon felt pledgeted horizontal mattress suture, suturing from the inside to the outside of the aorta. (Reprinted with permission from Parker et al, Ann Thorac Surg 1975;19:436–42)

View larger version (37K): [in this window] [in a new window]   Fig 14. Placement of a graft in the ascending aorta for dissection of the ascending aorta with a running horizontal mattress suture buttressed with pledgets followed by a continuous hemostatic whipstitch. (Reprinted with permission from Parker et al, Ann Thorac Surg 1975;19:436–42)

The authors carefully describe the technique of use of a heparin-coated vascular shunt for thoracic and great vessel procedures to eliminate the need for systemic heparin, to avoid hypertension during clamping of the aorta, and for perfusion of the distal circulation without having to utilize a pump (Figs 15 and 16). This experience included 25 patients operated on between 1966 and 1976. Twenty patients had resections of aneurysms of the descending thoracic aorta with one death among 15 patients undergoing elective resection. There were two deaths among five patients being operated on emergently. In addition there were five patients in whom the shunt was used who had procedures involving the great vessels. The authors reported no complications attributable to the shunt in either group and advocated this technique. This is one of the first reports demonstrating the efficacy of the Gott shunt.

View larger version (85K): [in this window] [in a new window]   Fig 15. Illustration of the technique for perfusing the lower half of the body with a Gott shunt for a descending thoracic aneurysm repair with the proximal cannulation being performed either through the left subclavian artery or the ascending aorta. (Reprinted with permission from Donahoo et al, Ann Thorac Surg 1977;23:507–13.)

View larger version (59K): [in this window] [in a new window]   Fig 16. Proximal shunt insertion is into the left ventricular (LV) apex and the distal cannulation can be either through the femoral artery, iliac artery, or the distal descending aorta. (Reprinted with permission from Donahoo et al, Ann Thorac Surg 1977;23:507–13.)

They reviewed their experience with multiple valve replacements over a 6-year period from 1970 to 1976, dividing their experience into two periods, before and after 1974. Sixty-two patients were operated on before 1974 with an operative mortality of 34% and 44 patients were operated on after 1974 with an operative mortality of only 11%, a two-thirds reduction. In the later time period ischemic arrest and topical hyperthermia were employed for myocardial protection as opposed to coronary perfusion during the earlier period. Cardiopulmonary bypass time was also reduced. In group II intraaortic balloon pumping also became available. The paper therefore is a good example of how technical progress was made over a period of time including better myocardial preservation and simplification of operative technique with improved results and supports the findings of Robicsek's earlier paper.

Craver, Jones, McKeown, Bone, Hatcher, and Kandrach [22] reported in 1981 at the Association meeting in Palm Beach on the Emory experience with porcine cardiac xenograft valves.

This was a very large series that included 1,093 patients with an operative mortality of 3.7% for aortic valve replacement, 7.8% for mitral valve replacement, and 4.7% for aortic valve replacement plus mitral valve replacement. The follow-up time was a maximum of 7.3 years with a mean of 1.89 years, with an actuarial survival for aortic valve replacement of 84% at 56 months, mitral valve replacement 84% at 56 months, and aortic and mitral valve replacement of 86% for 30 months (Fig 17). Only 0.7% of patients experienced thromboembolism; 4.8% of hospital survivors eventually had valve dysfunction (Fig 18); and there was a 1.7% late mortality and 3.1% of incidence of reoperation. There was an increased rate of valve dysfunction beginning at the sixth postoperative year and valve dysfunction was noted to be greater in patients who were younger than 34 years of age (Fig 19); and 3.1% of patients had a second valve replacement for late dysfunction. The authors concluded that porcine valves functioned well for up to 7 years with a low incidence of morbidity and mortality in patients more than 34 years of age. This was one of the earliest and largest reports on early and late results of porcine valvular prosthetic use and added considerable knowledge to the medical literature.

View larger version (23K): [in this window] [in a new window]   Fig 17. Overall actuarial survival curves are shown for patients who had aortic valve replacement (AVR, circles), mitral valve replacement (MVR, triangles), combined aortic and mitral valve replacements (AVR + MVR, squares), or aortic valve with coronary artery bypass (AVR + CAB, inverted triangles) with hospital mortality rate for each of those procedures. (Reprinted with permission from Craver et al, Ann Thorac Surg 1982;34:16–21)

View larger version (26K): [in this window] [in a new window]   Fig 18. The actuarial survival curves reflecting freedom from valve failure with patients older than 35 years of age with aortic valve replacement, mitral valve replacement, or combined aortic and mitral valve replacement are shown. (Reprinted with permission from Craver et al, Ann Thorac Surg 1982;34:16–21)

View larger version (18K): [in this window] [in a new window]   Fig 19. The effect of patient age on freedom from xenograft valve failure is demonstrated. (Triangles = >35 years; circles = 35 years or less.) (Reprinted with permission from Craver et al, Ann Thorac Surg 1982; 34:16 to 21)

The authors reported that from January 1972 through August 1977, 472 patients had internal mammary artery (IMA) coronary bypass, of which 100 were double mammary bypasses. The study selected only those patients who had a widely patent IMA 1 year postoperatively and then had a second catheterization 49 to 105 months after the operation (mean 64 months). These patients were not specifically recalled for this study but had follow-up catheterizations for multiple other reasons including other studies. All of the 91 left IMAs and 22 right IMA bypasses were patent at late catheterization but one right IMA was diffusely narrowed. There were 100 patent saphenous vein bypasses at 1 year and 87 at late catheterization. The authors note that first year closure of coronary bypass grafts had been 13.7% in their overall experience, with the saphenous vein grafts to the left anterior descending artery having a 10.3% 1-year occlusion rate and left IMA to the left anterior descending artery having a 5.7% closure rate. It was noteworthy that no internal mammary grafts totally failed during follow-up after being widely patent at 1 year. Thirteen saphenous vein grafts failed during the follow-up period, for a cumulative late patency of 81%. The authors, as part of this study determined that the overall patency for 285 left IMA grafts in patients having a 1-year or later catheterization in which narrowed grafts are considered to be occluded was 89.2%, and in their later experience it was 92.2%. Patency for 72 right IMA grafts using the same criteria was 87.5% and that did not improve with experience. It is interesting to note that vein graft patency was also excellent, being at 87% for this interval and at 81% overall. This paper was very important documentation of the value of internal mammary artery bypasses for long-term patency when many surgeons were still reluctant to adopt that conduit as their conduit of choice.

Urschel, Maruf, Razzuk, and Leshnower [24] reported on bypass grafting and aneurysmorrhaphy for aortic arch aneurysms at the meeting in Palm Beach, Florida, in 1981.

The authors described a technique where a size 26 low-porosity woven Dacron graft is sutured into the ascending aorta end-to-side and into the descending aorta end-to-side proximal and distal to the arch aneurysm without the use of cardiopulmonary bypass or systemic heparinization (Fig 20). This was done through a left posterolateral thoracotomy extending across the sternum transversely at the level of the fourth intercostal space. After insertion of this bypass, which will be permanent, a Crafoord clamp is placed obliquely across the inferior aspect of the arch aneurysm, preserving the aorta around the take-off of the arch vessels. A second clamp is placed on the descending aorta distal to the aneurysm. The aorta is divided there and oversewn and the aneurysm is opened just inferior to the Crafoord clamp, evacuated, and then also oversewn (Fig 21).

View larger version (47K): [in this window] [in a new window]   Fig 20. After lowering the systolic blood pressure chronologically to 90 mm Hg a partial occlusion clamp is placed on the ascending aorta and a Dacron graft is sutured in place end to side on the ascending thoracic aorta. (Reprinted with permission from Urschel et al, Ann Thorac Surg 1983;35:579–83).

View larger version (61K): [in this window] [in a new window]   Fig 21. The graft is then sutured into the descending aorta, again using a partially occluding clamp. The inferior aspect of the aortic arch aneurysm is stapled after opening it and removing the clot. The native ascending aorta supplies the brachiocephalic vessels. (Reprinted with permission from Urschel et al, Ann Thorac Surg 1983;35:579–83).

Livesay, Cooley, Reul, Walker, Frazier, Duncan, and Ott [25] reported on resection of aortic arch aneurysms, a comparison of hypothermic techniques in 60 patients, at the Association meeting in 1982 at Hilton Head Island.

The authors reviewed a 5-year experience of 60 patients operated on from November 1976 to February 1982 who underwent resection and reconstruction of the transverse aortic arch. In the first 20 patients operated before November of 1980 deep hypothermia (14°C to 18°C) was used and beginning in November of 1980 another 40 patients were operated on with modified techniques including moderate hypothermia for circulatory arrest (22°C to 26°C). A different method of preclotting the graft and only performing a single distal aortic anastomoses for arch reconstruction was implemented. In both periods the graft replacement was preformed with a low-porosity Dacron graft. In group I nonheparinized blood was used to preclot the grafts. In group II heparinized blood was drawn from the patient and centrifuged for 10 minutes, the plasma supernatant was separated from the red cells, and this was used to soak the graft. It was then placed into a steam autoclave for 5 minutes at 130°C (Table 13). Circulatory arrest was on average 22 minutes in group 1 and 12 minutes in group 2 with ranges from 8 to 46 minutes and 6 to 33 minutes respectively. The total period for cardiopulmonary bypass was significantly longer in group 1 patients (111 minutes versus 87 minutes), partly because of time required for rewarming of the patients who were cooled to a lower temperature. There was an operative mortality of 50% in the 20 patients in group I, half of which was attributed to uncontrolled hemorrhage. There was a significant improvement in survival in group II (90%) with no patient dying of a bleeding complication. Fixed neurologic deficits occurred in three patients in each group. An example of the simplified surgical method with a single distal anastomoses and patch reimplantation of the arch vessels is noted in Figure 22. This paper demonstrates the improved outcomes that can be achieved in very complex surgery by continuous refinement of techniques.

View larger version (22K): [in this window] [in a new window]   Fig 22. The usual technique for a patient with a type C aneurysm of the transverse aortic arch is illustrated, with the graft being inserted end to end proximal and distal to the arch and a large button of the aortic tissue at the takeoff of the brachiocephalic vessels being implanted into the superior aspect of the graft. (Reprinted with permission from Livesay et al, Ann Thorac Surg 1983;36:19–28).

The authors reported on 64 consecutive patients who had penetrating cardiac injuries treated between 1977 and 1983 at the University of Maryland Hospital. Twenty-eight had major associated injuries to other organs. The patients were divided into three groups according to their clinical status on arrival into the emergency department. Group 1 consisted of 13 patients who arrived pulseless without spontaneous movement. Group 2 included 43 patients who arrived who arrived in shock (blood pressure less than 90 mm Hg) due to tamponade or hemorrhage or both. Group 3 included eight patients who arrived conscious with a systolic blood pressure of 90 mm Hg or greater. Emergency room thoracotomy was utilized for lifeless or deteriorating patients. These patients required immediate cardiopulmonary bypass for repair of their injuries. Fifty-seven percent of the patients undergoing emergency room thoracotomy survived. Most of the deaths occurred, as one would suspect, among patients who arrived lifeless from gunshot wounds. Eighty-nine percent of the 27 patients who were stable enough to be transported to the operating room and corrected there survived. The overall survival in their group was 70%. None of the survivors had serious permanent neurologic complications. Three of their patients underwent late reoperation for closure of intracardiac defect and two required pericardiectomy. When examining survival in groups, group 1 (the lifeless on presentation) had a 23% overall survival, group 2 (those presenting in shock) a 79% survival, and group 3 a 100% survival. The authors attribute their success to rapid triage of patients in the emergency department, treating those in extreme distress there and transporting the others to a more controlled environment into the operating room.

At the 1984 meeting of the Association at Hilton Head Island, South Carolina, Myers, Davis, Foster, Maynard and Kaiser [27] reported on the surgical survival in the Coronary Artery Surgery Study (CASS) registry

The authors analyzed 8,991 patients in the registry portion of CASS who had a primary isolated CABG procedure and 8,971 who were followed up for more than 30 days. The 5-year survival for these patients was 90% and the operative mortality was 2.37% (Fig 23). Patients with left main coronary disease had an operative mortality of 3.84% and a 5-year survival of 85% as compared with patients without left main disease who had an operative mortality of 2.12% and a 5-year survival of 91% (Fig 24). Patients with single and double vessel disease had a somewhat higher long-term survival than those with three vessels or left main disease (Fig 25). Patients with relatively normal left ventricular function had a 5-year survival of 92% and an operative mortality of 1.97%. Those with moderate impairment had a 5-year survival rate of 80% and an operative mortality of 5.21%; and in those with poor ventricular function the 5-year survival was only 65% and operative mortality was 6.21% (Fig 26). This paper is a predecessor of many of the risk models used today using a stepwise cox logistic regression analyses. Twenty-nine variables were entered into the model and left ventricular wall motion score, congestive heart failure score, age, number of operated vessels, smoking history, left ventricular end-diastolic pressure, and percent of left human coronary artery stenosis were found to have a significant predictive value for long-term survival excluding 30-day mortality. These variables plus surgical priority and height influenced the surgical mortality. It is of interest that when height was used in the Cox proportional hazards model, the female sex was eliminated as a significant predictor of death. This study only included the first coronary bypass procedure if more than one was performed. This article was an incredibly high volume review which established many of the criteria for preoperative risk factors and benchmark outcomes for coronary bypass procedures.

View larger version (11K): [in this window] [in a new window]   Fig 23. Survival among 8,971 patients having a primary isolated coronary bypass procedure, excluding patients with valvular and ventricular procedures and those undergoing reoperation. Only the first coronary bypass operation was included. (Circles = all cases.) (Reprinted with permission from Meyers et al, Ann Thorac Surg 1985;40:245–60).

View larger version (17K): [in this window] [in a new window]   Fig 24. Survival among 1,303 patients with left main coronary artery stenosis 50% or more compared with 7,668 patients without left main stenosis (solid circles) meeting that criteria. The hollow circles indicate the survival of 104 patients with left main stenosis 50% or greater who had left dominate circulations. The triangles indicate left main stenosis patients or right dominant or balanced coronary circulations. (Reprinted with permission from Meyers et al, Ann Thorac Surg 1985;40:245–260).

View larger version (18K): [in this window] [in a new window]   Fig 25. The effect of the extent of coronary artery disease on surgical survival is demonstrated, with left main coronary artery (squares) disease having the worst long-term prognosis. (Open circles = one vessel; triangles = two vessel; solid circles = three vessel.) (Reprinted with permission from Meyers et al, Ann Thorac Surg 1985;40:245–60).

View larger version (18K): [in this window] [in a new window]   Fig 26. The effect of ejection fraction on long-term survival is shown, with a severe decrease in survival among patients with ejection fraction of less than 35% (open circles), a moderate decrease in survival among patients with an ejection fraction of 35% to 49% (triangles), and almost identical survival among patients with an ejection fraction of 50% to 69% (solid circles) and greater than 69% (squares). (Reprinted with permission from Meyers et al, Ann Thorac Surg 1985;40:245 to 60).

The authors studied the relationship of patient survival to HLA (A and B loci). In a review of 164 consecutive cyclosporine-treated patients who underwent heart transplantation between 1980 and 1986 at Stanford, all patients except those in whom HLA typing was not available were included. It was noted that there was a difference in actuarial 4-year survival with 0 to 1 mismatch having 88% survival; 2 mismatches, 70%; 3 mismatches, 59%; and 4 mismatches, 54% (Fig 27). However there was no difference in rejection rates (Table 14) depending on HLA compatibility but the patients with 3 or 4 mismatches had a significantly higher incidence of infections (Fig 28). The data therefore demonstrated that well-matched HLA grafts are associated with better long-term survival and fewer infections in cardiac transplant patients but did not affect rejection and was not related to doses of immunosuppressive drugs. HLA-DR typing was not performed. This paper, as many others, does not demonstrate a relationship between HLA compatibility and rejection in heart transplantation and is one of the first in the cyclosporin era to study this.

View larger version (14K): [in this window] [in a new window]   Fig 27. Actuarial survival according to the number of mismatches is shown for 0, 2, 3, and 4 mismatches. (Reprinted with permission from Frist et al, Ann Thorac Surg 1987;44:242–6).

View larger version (17K): [in this window] [in a new window]   Fig 28. The actuarial rate of infection by number of mismatches is shown. These data include all types of infection (bacterial, fungus, protozoan, and mycrobacterial) except for viral. (POSTOP = postoperative.) (Reprinted with permission from Frist et al, Ann Thorac Surg 1987:44;242–6).

Their series included 11 patients operated on beginning in 1982, nine of whom received a pneumatic Pierce-Donachy pump (Thoratec Laboratories) and two received an electrical Novacor pump. Seven patients required left ventricular support and four required biventricular support. Duration of support ranged from 8 hours to 91 days. In five patients contraindications to transplantation developed. Of the remaining six (four left ventricular assist device, two biventricular assist device) the three patients supported the longest were ambulatory while waiting a donor heart. All six patients underwent successful transplantation having been on the pumps from 8 hours to 91 days. Five of the six survived long-term with one late date due to noncompliance. This was an important paper because it demonstrated the effectiveness of ventricular assist devices as early as 1987 as a bridge to transplantation.

In 1987 at the 34th annual meeting of the Association, Vauthey, Berry, Snyder, Gilmore, Sundgarrd-Riise, Mills, and Ochsner [30] reported on their experience with left ventricular aneurysm repair combined with CABG, reviewing 246 consecutive patients older than 15 years.

The authors reviewed the 246 patients operated on from 1970 to 1985 and reported an overall operative mortality of 7.3%. Ten of the 18 deaths were cardiac related. Operative mortality was significantly impacted by mitral regurgitation, the mortality in that group being 22% versus 4.8% in those without mitral regurgitation. The overall 5-year survival rate was 69% (Fig 29). Most of the late deaths were caused by myocardial infarction. Significant predictors of decreased long-term survival included preoperative congestive heart failure (Fig 30), poor left ventricular function (Fig 31), an elevated left ventricular end diastolic pressure, New York Heart Association functional class IV (Fig 32), and the presence of mitral regurgitation (Fig 33). The most common perioperative complication was arrhythmia, which occurred in 24.4% of patients. This group tended to bypass the left anterior descending if it was a large sized artery with septal perforators or with diagonal branches and collateral blood flow. They did this in 51% of their patients. They noted a 5-year survival of 88% in patients who had the left anterior descending artery revascularized with an internal mammary graft compared with saphenous vein (72%) and no grafts (65%).

View larger version (9K): [in this window] [in a new window]   Fig 29. Survival for 246 patients having left ventricular aneurysm resection and repair with myocardial revascularization. (Reprinted with permission from Vauthey et al, Ann Thorac Surg 1988;46:29–35).

View larger version (11K): [in this window] [in a new window]   Fig 30. The difference in survival depending on presenting symptomatology is shown, comparing angina patients (heavy line; n = 105) with congestive heart failure patients (thin line; n = 102) and asymptomatic patients (dotted line; n = 24). Note that those with congestive heart failure fare less well. (Reprinted with permission from Vauthey et al, Ann Thorac Surg 1988;46:29–35).

View larger version (10K): [in this window] [in a new window]   Fig 31. The effect of preoperative ejection fraction on early and long-term survival after left ventricular aneurysm resection and repair is shown. Patients with an ejection fraction of less than 35 (solid line; n = 49) have a considerably worse survival rate than do patients with an ejection fraction of 35 or greater (dotted line; n = 40). (Reprinted with permission from Vauthey et al, Ann Thorac Surg 1988;46:29 to 35).

View larger version (13K): [in this window] [in a new window]   Fig 32. The difference in survival depending on New York Heart Association functional class is shown. Functional class IV patients (dashed line; n = 46) do much worse than class II (solid line; n = 88) and class III (dotted line; n = 84) patients. (Reprinted with permission from Vauthey et al, Ann Thorac Surg 1988;46:29–35).

View larger version (11K): [in this window] [in a new window]   Fig 33. The effect of mitral valve regurgitation in combination with the left ventricular aneurysm on survival is shown. Patients with mitral regurgitation (heavy line; n = 36) are at greater risk than patients with no regurgitation (light line; n = 210). (Reprinted with permission from Vauthey et al, Ann Thorac Surg 1988;46:29–35).

This report reviewed 95 heart transplants performed at the Johns Hopkins Hospital from 1983 to 1988. There were 19 deaths, 281 rejection episodes, and 180 distinct infections that were analyzed during a follow-up of 1 month to 62 months (Fig 34). The hazard function for rejection was biphasic, with a rapidly decelerating early phase during the first year followed by a constant late phase. By comparison the hazard function for infection was triphasic, with a delayed, decelerating early phase, a period of increased risk approximately 2 years after the operation, and then a constant late phase. (Fig 35). Nonsurviving patients had significantly higher rejection and infection rates during both the early and late phases compared with survivors, suggesting a complex interrelationship between infection and rejection as a determinate of late survival. This paper sheds further light on the timing of rejection and infection after heart transplantation, and the interrelationship between the two.

View larger version (17K): [in this window] [in a new window]   Fig 34. The total hazard ratio for death after cardiac transplantation is shown. Note the triphasic pattern with an early phase of increased risk followed by rapidly decelerating risk, an intermediate phase of resurgent risk, and a third light phase of low but increasing risk. (Reprinted with permission from Greene et al, Ann Thorac Surg 1989;47:650–4).

View larger version (16K): [in this window] [in a new window]   Fig 35. The hazard functions for death from all causes (circles), death from rejection (squares), and death from infection (triangles) are displayed after cardiac transplantation. (Reprinted with permission from Greene et al, Ann Thorac Surg 1989;47:650–4).

This was a large series and included patients operated on after 1976 with 975 receiving porcine mitral valve replacements, 169 receiving mechanical mitral valve replacements, and 280 receiving Carpentier-type mitral valve reconstruction. The overall operative mortality was 5% for mitral valve reconstruction, 16.6% for mechanical valve replacement, and 10.6 for porcine valve replacement. Five-year survival including the hospital deaths was 76% for valve repair, 72% for mechanical mitral valve replacement, and 69% for porcine valve replacement (Fig 36). The predicators of increased operative risk and decreased survival were increasing age, New York Heart Association functional class IV status, previous cardiac operation, and performance of concomitant cardiac surgical procedures. The type of valvular procedure was not predictive of operative risk or overall survival. However the 5-year freedom from reoperation was 94% for nonrheumatic patients having reconstruction, 77% for rheumatic patients having reconstruction, 96.4% for mechanical replacement, and 96.6% for porcine replacement. The 5-year freedom from all valve related morbidity and mortality was significantly better for valve reconstruction compared with both types of valve replacement (Fig 37). This large series further demonstrated the efficacy of mitral valve reconstruction and its durability in nonrheumatic mitral valve disease.

View larger version (22K): [in this window] [in a new window]   Fig 36. Actuarial survival from all cardiac-related deaths after mitral valve repair and replacement. (Circles = repair; squares = mechanical valve; triangles = porcine valve; POST-OP = postoperative.) (Reprinted with permission from Galloway et al, Ann Thorac Surg 1989;47:655–62).

View larger version (23K): [in this window] [in a new window]   Fig 37. Actuarial freedom from all cardiac-related morbidity and mortality after mitral valve repair and replacement. (Circles = repair; squares = mechanical valve; triangles = porcine valve; POST-OP = postoperative.) (Reprinted with permission from Galloway et al, Ann Thorac Surg 1989;47:655–62).

The authors reported on a model derived from a Bayesian analysis of 6,630 patients undergoing coronary bypass in the Coronary Artery Surgery Study (CASS). Using this model they found that age, sex, ventricular function, previous myocardial infarction, and the extent of coronary disease, unstable angina, and surgical priority were significant predictors of operative mortality and were used to sort patients into risk categories. They then examined the results at their own institution, the Walter Reed Army Medical Center, from January 1984 through December 1987, reviewing 840 patients who underwent isolated coronary bypass. They noted that the unadjusted operative mortality at Walter Reed was 3.9%, which was higher than the 2.3% reported in the CASS registry (p < 0.001). However when the patients were entered into the CASS model for risk stratification the mortality conformed to the CASS experience (Table 15). The authors make the strong argument that this is a good example of why risk adjusted outcomes need to be used for institutional comparisons rather than unadjusted outcomes and that this is a fundamental principal of quality assurance. This was one of the very first papers advocating the use of risk-adjusted outcomes as a quality improvement tool.

They reviewed 1,298 patients who had received this prosthesis at their institution over a 10-year period from October 1977 to October 1987, which marked the 10-year anniversary of the availability of that valve on the market. They experienced an early mortality of 5.7%. Ninety-three percent complete follow-up was accomplished. A late mortality of 16.9% occurred (Fig 38). Sixty-four patients had thromboembolic episodes, 24 anticoagulant related bleeding, four had valve thrombosis, seven had prosthetic endocarditis, nine had a perivalvular leak, and 10 underwent reoperation (Table 16). There were no structural valve failures. The total number of valve related deaths was 37. Linearized rates for thromboembolism, valve thrombosis, and anticoagulant related bleeding were 1.49%, 0.09%, and 0.56% patient years respectively. The actuarial incidence of freedom from all complications was 66.9%. Valve-related death was 0.9% at 9 years. Further analysis revealed that more than 50% of the deaths were due to ongoing arteriosclerosis or myocardial issues unrelated to the prosthesis. Only 3.2% of the patients died secondary to prosthetic valve complications. This study helped to substantiate the safety durability and efficacy of the St. Jude mechanical prostheses and had a very high rate of follow-up.

View larger version (29K): [in this window] [in a new window]   Fig 38. Actuarial survival curves for the entire group of patients (A) including operative and late deaths and (B) excluding operative deaths. (AVR = aortic valve replacement [circles]; DVR = double-valve replacement [Xs]; MVR = mitral valve replacement [squares].) (Reprinted with permission from Arom et al, Ann Thorac Surg 1989;47:831–7)

These patients were operated on between 1976 and 1989. The operative technique of the composite valve-conduit with coronary implantation is described in detail (Figs 39 and 40). Of interest is that the authors originally performed a standard Bentall procedure with utilization of the Cabral procedure in some patients when the coronary ostea were low but more recently they have incised the left and right coronary ostea from the aortic wall to mobilize them, allowing adequate length to bring them up superiorly to where they reach without tension into the graft. Twenty-two patients had ascending aortic dissection at the time of composite graft repair and 18 also had a mitral valve procedure. Among the 92 patients undergoing elective repair there were no hospital deaths. One of eight patients who underwent an emergency repair of a ruptured aneurysm died in the operating room. The overall hospital mortality was 1% and there have been 10 late deaths among the 99 survivors for a 10.1% late death rate, with only five deaths in the last 88 patients (5.7%). In six patients late endocarditis developed, resulting in three of the late deaths. Three others had this treated by a re-replacement of the aortic root with a cryopreserved homograft. Actuarial survival for the 100 patients was 92.6% at 5 years and 75.8% at 10 years (Fig 41). The authors conclude that composite graft repair for Marfan's aneurysms of the ascending aorta can be performed safely with good long-term results. In addition based on their own experience and that of others in the literature, they believe that aneurysms 6 cm or greater in diameter should be repaired even in the asymptomatic patient. This is truly a remarkable study because of its size and superb early and late results for this very complex problem.

View larger version (88K): [in this window] [in a new window]   Fig 39. The technique of standard composite graft insertion is illustrated A through H for repair of ascending thoracic aneurysms in Marfan syndrome patients . How sutures are placed below the annulus when the coronary ostea are low is shown in E. In I, note that the aorta is completely transected to facilitate the end-to-end anastomosis. Also a no-wrap technique is used with tacking loosely the redundant aneurysm wall over the graft. (a. = artery; L.V. = left ventricular; R. = right; RA = right atrium; r.sup.pulm.v. = right superior pulmonary vein; RV = right ventricle.) (Reprinted with permission from Gott et al, Ann Thorac Surg 1991;52:38–45).

View larger version (165K): [in this window] [in a new window]   Fig 40. The Cabrol technique for graft interposition is demonstrated. Note that this technique was replaced by the authors by directly mobilizing the low lying coronary ostea as shown and then directly anastomosing them end to side into the graft. (A = artery; L = left; R = right.) (Reprinted with permission from Gott et al, Ann Thorac Surg 1991;52:38–45).

View larger version (12K): [in this window] [in a new window]   Fig 41. Actuarial survival of the 99 patients discharged from the hospital after the composite graft repair. The number of patients at risk at each time is shown. (Reprinted with permission from Gott et al, Ann Thorac Surg 1991;52:38–45).

The authors reviewed the medical records of 93 consecutive patients who were admitted with penetrating injury of the aortic arch and its branches to the Elvis Presley Trauma Center in Memphis; 63.5% of those injuries were gunshot wounds. The authors operated immediately when there was evidence of massive or pulsatile bleeding or persistent hemodynamic instability. Patients in stable condition who had a pulse deficit, major hematoma, widened mediastinum, or entrance or exit wounds thought to be near any major arteries had aortograms for identification of potential injuries and also the localization of these injuries. The authors noted that having operating rooms and arteriographic suites in the trauma center was very helpful.

This paper describes in detail the techniques utilized for the treatment of these injuries, which are in very difficult anatomic locations for surgical control. The most common incisions employed were midline sternotomy with cervical extensions over one or both carotids when needed or an anterior left thoracotomy through the fourth intercostal space (with dislocation of the cartilage) for control of the left subclavian artery and then extension of this incision to include on occasion excision of part of the clavicle for exposure of the distal left subclavian. Intermittent aortic occlusion, either partial or complete, and caval occlusion were useful adjuncts for exposure and decrease in blood loss and for decreasing blood pressure for suturing. The authors also utilized shunts from the ascending aorta to the carotid arteries when injuries occurred at the origin of the carotid or innominate arteries from the aorta, obtaining proximal control intrapericardially and distal control on the neck before opening the hematoma. Occasionally permanent grafts were inserted to bypass injuries. Only one patient required cardiopulmonary bypass. The authors report 41% survival for aortic injuries, 83% for innominate artery injuries, 81% for carotid injuries, and 86% for subclavian injuries. Patients who presented with a neurologic deficit did not get better nor did those who came in without a neurologic deficit experience one. Thirteen percent of the patients could not be resuscitated to get to the point of operative repair. Overall survival was 71%. This was a very large and detailed review of how to best manage these very difficult and challenging injuries.

Safi, Brien, Winter, Thomas, Maulsby, Doerr, and Svensson [37] reported on their experience with brain protection through cerebral retrograde perfusion during aortic arch aneurysm repair at the 1992 meeting of the Association in Wesley Chapel, Florida.

The authors reported on 11 patients who underwent resection and graft replacement of ascending and aortic arch aneurysms and describe a technique for retrograde perfusion of the brain. This perfusion was performed at 15°C to 24°C and was administered through the superior vena cava. The mean cerebral ischemic time was 35 minutes, ranging from 11 to 71 minutes, and throughout the ischemic period they performed retrograde perfusion, with a sampling of innominate and left carotid artery blood at 1 and 5 minutes and every 10 minutes thereafter for analysis of arterial oxygen content. All patients survived. All except one awoke neurologically intact; in that patient there was evidence before circulatory arrest of injury. They also measured creatine kinase BB but concluded that it was not a good marker to detect brain injury.

The authors cardiopulmonary bypass cannulation technique was right femoral artery and vein with immediate cooling. When cooling was complete the chest was entered through a median sternotomy. The superior vena cava was cannulated and a left ventricular vent was placed into the right superior pulmonary vein with tourniquets placed around the cavae. When the temperature reached the desired level of hypothermia the patient was placed in the Trendelenburg position and circulatory arrest was initiated. The superior vena cava cannula was connected to an arterial line from the heart/lung machine and retrograde cerebral perfusion was immediately started, keeping the pressure less than 25 mm Hg. The aneurysms were then repaired. After completion of the aneurysm repair and with the patient still in head-down position the flow was restarted by the femoral artery to flush out air and debris. A sidearm of the graft to the transverse aortic arch was connected to an arterial line of the heart/lung machine and the flow started in an antegrade fashion to begin rewarming. Ten of the 11 patients awoke neurologically intact. The mean time to wake-up was 4 hours. Both patients with circulatory arrest times of greater than 70 minutes woke within 6 hours. One patient had a stroke, probably due to emboli. Although this was a small series it very meticulously describes the techniques for retrograde cerebral perfusion with excellent results.

In 1993 Jones, Hill, Vasilakis, Hollingsed, Graeber, Gustafson, Cruzzavala, and Murray [38] reported on experience with heparin-coated bypass circuits incorporated into the cardiopulmonary bypass system.

The authors studied 20 patients in a high-risk subset of patients who had a history of recent stroke, posttraumatic injuries, recent gastrointestinal bleeding, protamine allergies, combined cardiac and noncardiac procedures, and ventricular assist devices. They used this technique keeping the heparin dosage minimal, maintaining the activated clotting time at 200 seconds, and noted that the postoperative blood loss and transfusion requirements were minimal. The postoperative complement levels of C3a and C5a were normal, suggesting a good biocompatibility and a decreased inflammatory reaction. They concluded that the use of this type of bypass circuit could benefit older patients with preoperative pulmonary dysfunction, as well as the type of high-risk patients included in the study. This was a careful study of the newer heparin-coated bypass circuits, which may offer some protection against the development of an inflammatory reaction commonly seen with cardiopulmonary bypass.

In 1994 Craver, Bufkin, Weintraub, and Guyton [39] reported the incidence of neurologic events after coronary bypass grafting with warm cardioplegia at the 41st annual meeting at Marco Island.

They noted that previous reports had shown that warm heart surgery and warm cardioplegia offered a good myocardial protection benefit but there was an added neurologic risk. Craver and associates hypothesized that moderate hypothermic aerobic heart surgery would maintain the myocardial protection and reduce neurologic risk. They compared 493 patients undergoing coronary artery bypass graft operations with normothermic (35°C to 37°C) continuous blood cardioplegia and normothermic perfusion and 379 patients undergoing the same operation with hypothermic (33°C to 29°C) continuous blood cardioplegia and hypothermic perfusion to test this hypothesis. The postoperative myocardial infarction rate was 1.2% in the hypothermic group and 1.3% in the normothermic group. The requirement of an intraaortic balloon pump was 3.4% in the hypothermic group compared with 1.4% in the normothermic group. However the incidence of postoperative neurologic events was 4.7% in the normothermic group versus 1.8% in the hypothermic group (p = 0.038; Table 17). When examining this using a multivariate analysis predicting stroke, older age and normothermic cardioplegia were significant predicators. The authors concluded that hypothermic continuous blood cardioplegia provided equivalent myocardial protection and lessened neurologic risks. This was an important study demonstrating the protective effects of hypotherm cardiopulmonary bypass for neurologic adverse outcomes.

This was a retrospective study from 1983 to 1993 of 96 patients with an ejection fraction of 0.25 or lower who underwent CABG. The hospital mortality was 8%. All patients had symptoms of heart failure. The average age was 63 patients with valvular heart disease other than mild to moderate, and those requiring a ventricular aneurysm resection or emergency operations for acute coronary occlusion were excluded. Catheterization films were reviewed by a blinded cardiologist. The authors found that increased age and poor vessel quality were the only significant predictors of poor outcomes. This was in contrast to sex, presence or absence of angina, degree of angina, preoperative ejection fraction, preoperative arrhythmia disorder, aortic cross-clamp time, and the number of bypass grafts, which had no significant effect on the outcome. The authors concluded that poor vessel quality and older age are predictors of poor outcome in patients with low ejection fraction undergoing coronary bypass and that poor distal coronary quality is a contraindication to CABG in these patients even when angina is present. This is one of the few studies available that evaluates the quality of distal coronary targets and analyzes the effect of poor run-off on outcomes.

In 1996 at the 43rd annual meeting of the Association, Hammon, Stump, Kon, Cordell, Hudspeth, Oaks, Brooker, Rogers, Hilbawi, Coker, and Troost [41] reported on risk factors and solutions for the development of neurobehavioral changes after CABG.

Between 1991 and 1994, 395 patients undergoing CABG were evaluated with an 11-part neurobehavioral battery administered preoperatively at 1 and 6 weeks postoperatively. In addition they were instrumented with continuous-wave carotid Doppler transducers intraoperatively to estimate cerebral microembolism. They correlated this data with surgical technical maneuvers during the operation and analyzed their results after 2 years, namely in the middle of the study. They used these data to influence surgical technique in the later 2 years. They found that increased patient's age, more than 100 emboli per case, and a palpable aortic plaque were significant predictors of patients developing new neurobehavioral deficits. Furthermore after adjusting techniques the more recent group of patients had a lower incidence of neurobehavioral impairment, 60% versus 69% at 1 week. At 1 month 29% of patients in each group demonstrated impairment (Fig 42). Different techniques added to group B patients included increased use of single cross-clamp technique, increased venting of the left ventricle, and application of transesophageal and epiaortic ultrasound scanning to locate and avoid trauma to aortic atherosclerotic plaques. This paper is important because it carefully assessed neurologic sequelae after bypass and the effects of operative techniques to decrease the incidence of neurologic dysfunction.

View larger version (15K): [in this window] [in a new window]   Fig 42. Percentage of patients having neuropsychologic impairment after coronary artery bypass graft surgery (n = 395) is compared between group A (open bars) and group B (shaded bars) both at 5 to 7 days and at 1 month. Group B has a borderline statistically significant increase in the earlier period and a significant decrease at the 1-month period. (Reprinted with permission from Hammon et al, Ann Thorac Surg 1997;63:1613–7).

The authors reviewed the 13-year experience at the University of Oklahoma of treating infective endocarditis of the aortic valve and root. During that period 81 patients underwent valve replacement for endocarditis, 78% of these patients having active endocarditis at the time of the operation. Thirty-six percent had prosthetic endocarditis. Homografts were used in 46 of the valve replacements, autografts (Ross procedure) in 25 patients, and prosthetic valves in 10. Eleven patients who had either a Ross procedure or homograft required mitral valve repair as well. Follow-up was 90% at 2 years with a mean of 3.7 years. Early mortality was 16% overall, 12% for the Ross procedure, 17% for homografts, and 20% for prosthetic aortic valve replacement Overall late mortality was 10% and the valve-related late mortality was 7%. Actuarial survival at 5 years was 88% in the Ross group, 69% in the homograft group, and 29% in the prosthetic group. Postoperative endocarditis occurred in 12.5% of patients with prosthetic valves and 3% of those in the homograft and Ross procedure groups. The authors concluded that autograft and homograft valve conduit replacements are preferred for endocarditis patients and that the risk of valve related death or recurrent endocarditis was low in that group of patients. They acknowledged that this was not a prospective randomized study and therefore there were differences in the patients receiving these different valves.

Franga, Kratz, Crumbley, Zellner, Stroud, and Crawford [43] presented early and long-term results of CABG in dialysis patients at the 1999 meeting of the Association in Puerto Rico.

Their study was a retrospective review of included 44 dialysis patients undergoing coronary bypass from 1984 to 1997. They compared their results with the results of overall coronary bypass in the Society of Thoracic Surgeons (STS) database. They found that their coronary bypass patients on dialysis had a 73% complication rate compared with 36% of all comers in the STS database. Postoperative cerebral vascular accidents occurred in 7% versus 1.7% in the "normal" STS coronary artery bypass graft surgery population and cardiac arrest in 7% versus 1.5%, both of which were statistically significant. The overall operative mortality was 11.4% in dialysis patients. Preoperative ejection fraction was predictive of operative mortality. No other risk factors were significant predictors of poor outcomes but the sample size was small.

The authors also captured quality of life, walking distances, and quality of distal coronary arteries. They noted that the angina class decreased from 2.8 to 1.5, and congestive heart failure class from 2.6 to 1.8 postoperatively. Quality of life scores did not improve but walking distances in late follow-up remained consistently improved. Actuarial survival at 5 years was only 32% (0% for smokers and 83.6% for nonsmokers; Figs 43 and 44). Distal diffuse disease was found to be a predictor of operative mortality. This study is important because it not only demonstrates the very poor 5-year survival of dialysis patients after coronary artery bypass but also the symptomatic improvement they experience.

View larger version (15K): [in this window] [in a new window]   Fig 43. Actuarial survival for all patients in the study is shown, demonstrating the mean number of patients that remain at each time and interval. (Post op = postoperative.) (Reprinted with permission from Franga et al, Ann Thorac Surg 2000;70:813–8).

View larger version (19K): [in this window] [in a new window]   Fig 44. Actuarial late survival is shown in smokers (circles) versus nonsmokers (squares). The five operative deaths are excluded and then again an n shows the number of patients remaining. (Post op = postoperative.) (Reprinted with permission from Franga et al, Ann Thorac Surg 2000;70:813–8).

This was a large single-hospital study of 238 survivors aged 17 to 82 years who underwent operation between 1986 and 1999. All procedures were done as root replacements. There were 145 autografts and 93 homografts utilized. The homograft group was a higher risk group and older (49 versus 35 years of age), had more previous aortic valve replacements (34% versus 8%), and had more active endocarditis at the time of the current operation (27% versus 7%). The maximum follow-up was 12 to 13 years for both groups. Late survival at 10 years was 77% for autografts and 67% for homografts (p = 0.13; Fig 45). Freedom from valve degeneration at 10 years was 97% in the autograft group versus 79% in the homograft group (p = 0.63; Fig 46). There was no statistical difference in the freedom from valve related complications or reoperations (Figs 47 and 48) although part of this may be a sample size issue since there was a 10 to 15 percentage point difference favoring the autograft. The authors acknowledge that it is difficult to compare the groups because of the higher preoperative risk factors in the homograft group but both valves have comparable late survival and the incidence of valve degeneration was low for both groups until about 8 years at which point there appeared to be a trend toward an advantage for the autograft over the homograft group, suggesting a benefit for younger patients. The authors thought that the autograft did not offer a significant advantage over the homograft for patients older than 60 years. This was one of the largest series ever reported of homograft and autograft aortic valve replacement comparing the early and late results of each.

View larger version (15K): [in this window] [in a new window]   Fig 45. This Kaplan-Meier graph depicts late survival after homograft (dashed line) or autograft (solid line) replacement of the aortic valve. The mean ± 67% confidence limit for the three time intervals is indicated. (Reprinted with permission from Knott-Craig et al, Ann Thorac Surg 2000;69:1327–31).

View larger version (16K): [in this window] [in a new window]   Fig 46. This Kaplan-Meier graph depicts valve degeneration of both the homografts (dashed line) and autografts (solid line) after replacement of the aortic valve. The mean ± 67% confidence limit for the three time intervals is indicated. (Reprinted with permission from Knott-Craig et al, Ann Thorac Surg 2000;69:1327–31).

View larger version (17K): [in this window] [in a new window]   Fig 47. This is a Kaplan-Meier graph demonstrating freedom from all valve-related morbidity and mortality after both homograft (dashed line) and autograft (solid line) replacements of the aortic valve. The mean ± 67% confidence limit for the three time intervals is indicated. (Reprinted with permission from Knott-Craig et al, Ann Thorac Surg 2000;69:1327–31).

View larger version (17K): [in this window] [in a new window]   Fig 48. This Kaplan-Meier graph illustrates freedom from all valve-related reoperations after homograft (dashed line) or autograft (solid line) replacement of the aortic valve. The mean ± 67% confidence limit for the three time intervals is indicated. (Reprinted with permission from Knott-Craig et al, Ann Thorac Surg 2000;69:1327–31).

This paper reviewed 22 consecutive patients who had a septomyectomy between 1981 and 1999. Preoperatively all patients had significant cardiac symptomatology with a mean resting gradient of 78 mm Hg. Nine of the patients required associated cardiac procedures at the time of their myectomy. There were no perioperative deaths and minimum morbidity, with two late deaths at 6 and 9 years postoperatively of congestive failure and arrhythmias. The other survivors have been followed up an average of 6.6 years and currently have minimal or no symptoms and a mean resting gradient of 12 mm Hg. This is a nice series with excellent results including excellent reduction in the mean resting gradient.

John, Rajashinghe, Chen, Weinberg, Sinha, Mancini, Naka, Oz, Smith, Rose, and Edwards [46] reviewed the long-term outcomes after cardiac transplantation at Columbia University and presented this at the annual meeting in 2000.

This was a very large retrospective review of 1,086 consecutive heart transplant recipients who underwent operation between 1977 and 1999. The patients were groups by four eras based on the type of immunosuppressive therapy: era I, steroids and azathioprine (1977 to 1983); era II, steroids and cyclosporine (1983 to 1985); era III, cyclosporine, steroids, and azathioprine (1985 to 1995); and era IV, cyclosporine, steroids, mycophenolate, and mofetil (1996 to 1999). The actuarial survival for the entire group was 79%, 66%, and 49% at 1, 5, and 10 years respectively, with differences in each of the eras demonstrated in Figure 49. Patients' age increased as did female sex in eras III and IV and early mortality and long-term survival improved significantly over all eras. Rejection decreased as a cause of death over time, being 24%, 21%, 15%, and 9% respectively in the four eras but transplant coronary artery disease as a cause of death remained unchanged. The authors concluded that cardiac transplantation provides satisfactory long-term survival and that survival has improved as immunosuppressive therapy has improved. Of note is that the overall early mortality rate was 10.3% but was 27%, 19%, 11%, and 7% in eras I, II, III, and IV respectively. Multivariate analysis showed that recipient age (Fig 50), female sex, a pretransplant diagnosis of congenital heart disease, and transplantation during eras I and II were significant predictors of operative mortality (Table 6). This is one of the largest single-institution reviews of short- and long-term outcomes after heart transplantation and demonstrates markedly improved results over time.

View larger version (25K): [in this window] [in a new window]   Fig 49. Actuarial survival of patients undergoing primary cardiac transplantation during 1977–1999. The squares represent actual events, positioned along the horizontal axis at the time of the event and by the Kaplan-Meier method along the vertical axis. (Reprinted with permission from John et al, Ann Thorac Surg 2001;72:440–9).

View larger version (31K): [in this window] [in a new window]   Fig 50. The actuarial survival of patients undergoing cardiac transplantation is shown by age group stratification. The squares (age younger than 18 years), triangles (18 to 60 years), and circles (more than 60 years) represent actual events as they occur along the horizontal axis. (Reprinted with permission from John et al, Ann Thorac Surg 2001;72:440–9).

Coronary bypass procedures from the database were reviewed for the calendar years 1998 and 1999 from 26 centers that had performed at least 20 off-pump procedures. During that period 118,140 total coronary bypass procedures were in the database and the number of off-pump cases was 11,717 (9.9%). It was noted that the off-pump patients were older, more likely to be female, and less likely to be diabetic than the conventional on-pump coronary artery bypass graft group. The off-pump group had a greater percentage of patients with preexisting chronic obstructive pulmonary disease, dialysis dependent renal failure, and cerebral vascular disease. The on-pump group had a greater proportion of patients with three-vessel coronary disease and more that were classified as urgent or emergent cases. In addition they received more bypasses. Risk-adjusted mortality in the on-pump group was 2.93% as compared with 2.31% in the off-pump group (p < 0.0001). The absolute reduction in risk adjusted mortality was 19% and for complications was 23%, the respective O/E ratios being 0.81 and 0.77 favoring the off-pump group (Figs 51 and 52). Among patients who had preexisting renal failure as defined by a creatinine of more than 1.5 there was no statistical difference between off-pump and on-pump in postoperative renal failure. However of those who had a history of cerebral vascular disease, the on-pump group had a 4.6% incidence of postoperative cerebrovascular attacks as compared with 2.5% of the off-pump group. In patients with chronic obstructive pulmonary disease, 11.3% of the on-pump group had postoperative ventilator dependence as compared with 8.9% of the off-pump group. The authors concluded that off-pump coronary bypass was associated with decreased mortality and morbidity and might prove superior in appropriately selected patients, realizing that this was a retrospective study and subject to patient selection bias. This study reported the largest review of off-pump verses on-pump coronary bypass and suggests that there may be a decrease in risk adjusted operative mortality and morbidity. However long-term outcomes and patency rates need to be determined in a prospective study to evaluate this recently developed technique.

View larger version (66K): [in this window] [in a new window]   Fig 51. Observed to expected ratios for operative mortality and morbidity are shown for off-pump and on-pump coronary bypass groups. There was a significant reduction in both mortality and morbidity with the off-pump group. (Reprinted with permission from Cleveland et al, Ann Thorac Surg 2001;72:1282–9).

View larger version (50K): [in this window] [in a new window]   Fig 52. Observed to expected risk ratios for various complications are shown. There was significantly less bleeding, renal failure, prolonged ventilation, and cerebrovascular attacks (CVA) in the off-pump group (Off) than on-pump group (On) after adjusting for patient risk. (Reprinted with permission from Cleveland et al, Ann Thorac Surg 2001;72:1282–9

The patients were operated on from 1992 to 1997 as part of a multicenter, worldwide investigation of the Medtronic Freestyle stentless porcine bioprosthesis. All patients received an aortic root replacement with reimplantation of the coronaries. The operative mortality was 3.9%. There were no structural valve deteriorations, nonstructural deteriorations, perivalvular leaks, or unacceptable hemodynamic performance. Survival at 8 years was 59.8% (Fig 53); freedom from thrombolytic complications was 83.3%, freedom from postoperative endocarditis, 96.9%, and freedom from reoperation, 100%. Systolic gradients were consistently between 6 to 7 mm Hg at 1 year and 8 years, with an effective orifice area of 1.9 at 1 year and 1.8 at 8 years. Ninety-eight percent of patients had either no or trivial aortic regurgitation at 1 year and 2% had mild aortic regurgitation. At 8 years 100% of patients evaluated were free of any aortic insufficiency (Fig 54). Of interest is that freedom from thromboembolic complications at 8 years was only 83%. The authors thus demonstrated excellent early and late results using this more complex technique of aortic valve replacement.

View larger version (16K): [in this window] [in a new window]   Fig 53. Actuarial survival after aortic root replacement is shown for an 8-year period. The open circles represent the estimated survival for 72-year-olds in the United States in 1997 from the National Vital Statistics Report, volume 47, December 13, 1999. The vertical lines are the 95% confidence intervals of survival for the free-style patients. (Reprinted with permission from Konn et al, Ann Thorac Surg 2002;73:1817–21).

View larger version (28K): [in this window] [in a new window]   Fig 54. The incidence of aortic regurgitation is shown at discharge and at 1, 4, and 8 years after aortic root replacement. (Solid bars = none/trivial; heavy hatched bars = mild; medium hatched bars = moderate; light hatched bars = severe.) (Reprinted with permission from Konn et al, Ann Thorac Surg 2002;73:1817–21).

The authors reviewed their experience from 1995 to 1999, which included 9,965 patients of whom 2,891 had diabetes and underwent isolated CABG. The diabetic and nondiabetic groups were then divided by whether they had their coronary bypass on pump or off pump. Twelve percent of each group underwent coronary artery bypass without cardiopulmonary bypass. The remainder were operated on using the pump. Nineteen preoperative variables were compared among treatment groups by univariate analysis. In general patients undergoing CABG without pump had a higher mean predicted mortality: diabetic patients 3.69% versus 3.72% (p = 0.83), nondiabetic patients 3.03% versus 2.86% (p = 0.79). In the nondiabetic group the patients in the off-pump group had a survival benefit compared with the on-pump group, with actual and risk-adjusted mortalities of 1.81% versus 3.44% (p = 0.0127), and 1.79% versus 3.61% (p = 0.007) respectively. The off-pump survival benefit was not seen in the diabetic patients whose unadjusted mortality was 2.89% versus 3.69% (p = 0.452), and risk-adjusted mortality, 2.19% versus 2.98% (p = 0.42). However the diabetic patients that received off-pump CABG had fewer complications including decreased blood product usage (34.39% versus 58.4%, p = 0.001), reduction of prolong ventilation (6.94% versus 12.1%, p = 0.005), less atrial fibrillation (15.90% versus 23.26%, p = 0.002) and less renal failure requiring dialysis (0.87% versus 2.75%, p = 0.036). The authors concluded that there was no survival advantage in diabetic patients having the CABG performed off pump but there was a significant reduction in morbidity. There was a survival advantage for the nondiabetic patient with CABG being performed off pump in both actual and risk-adjusted mortality.

Comment

The history of adult cardiac surgery as seen through the eyes of the membership of the Southern Thoracic Surgical Association as been presented. The papers reviewed a span of 50 years, beginning in 1954. These included the earliest very simplistic perfusion technology, carried us through evolution of myocardial revascularization beginning with the Vineberg procedure and extending into direct coronary bypass with various conduits. In addition methodologies have greatly changed for myocardial preservation and the risks of these procedures have been greatly reduced. Progress in valvular surgery has also been documented with the earliest reports of prosthetic valve replacement, review of homograft and autograft valve replacement, endocarditis, and hypertrophic cardiomyopathies. There were also a number of papers on the subject of aortic dissection and aneurysms with new technologies including retrograde cerebral perfusion during deep hypothermic circulatory arrest. Several trauma papers were also reviewed, indicating continuous improved results in the handling of major cardiac and great vessel trauma. One of the first papers on adult cardiac transplantation was reviewed as were several others in this field.

A common aspect of the Association has been the open and lively discussions that occur at the meetings, where the presenters fully expect a gamut of questions and expect to be challenged on their presentations. This has built a camaraderie that is somewhat unique to the Southern Thoracic Surgical Association; it reflects the ideals of intellectual honesty while seeking further knowledge in the profession and at the same time maintaining a sense of humor and humility.

Acknowledgments

The authors wish to thank Ms Kesinee, Ms Judi Arias, Ms Kami Fugate, and Ms Susan Abenilla-Brown for their efforts in preparing this manuscript.

References

1. Seiler H.H. The early years of the Southern Thoracic Surgical Associations. Ann Thorac Surg 1972;14:113-122.[Free Full Text]
2. Lewis F.J., Taufic M. Closure of atrial septal defects with the aid of hypothermia: experimental accomplishments and the report of one successful case. Surgery 1953;33:52.[Medline]
3. Lillehei C.W., Cohen M., Warden H.E., Varco R.L. The direct-vision intracardiac correction of intracardiac anomalies by controlled cross circulation. Surgery 1955;38:11.[Medline]
4. Gibbon J.H. Application of mechanical heart lung apparatus to cardiac. Surgery. Minnesota Med 1954;37:171.
5. Alford W.C. The heritage of the Southern Thoracic Surgical Association. A presidential address. Ann Thorac Surg 1997;63:305-308.[Free Full Text]
6. Sealy WC, Brown IW, Young WG, Stephen CR, Harris JS, Merritt D. Hypothermia, low flow extracorporeal circulation and controlled cardiac arrest for open heart surgery. Surg Gynecol Obstet 1957;104:441–50
7. Sewell W.H. Results of 122 mammary pedicle implantations for angina pectoris. Ann Thorac Surg 1966;2:17-30.[Free Full Text]
8. Yeh T.J., Anabtawi I.N., Cornett V.E., White A., Stern W.H., Ellison R.G. Bacterial endocarditis following open-heart surgery. Ann Thorac Surg 1967;3:29-36.[Free Full Text]
9. Cooley D.A., Bloodwell R.D., Hallman G.L., Nora J.J. Transplantation of the human heart. JAMA 1968;205:67-74.
10. Takaro T., Dart C.H., Scott S.M., Fish R.G., Nelson W.M. Coronary arteriography. Ann Thorac Surg 1968;5:213-215.[Free Full Text]
11. Diethrich E.B., Liddicoat J.E., Alessi F.J., Kinard S.A., DeBakey M.E. Serum enzyme and electrocardiographic changes immediately following myocardial revascularization. Ann Thorac Surg 1968;5:195-203.[Free Full Text]
12. Johnson W.D., Flemma R.J., Lepley D. Direct coronary surgery utilizing multiple-vein bypass grafts. Ann Thorac Surg 1970;9:436-444.[Free Full Text]
13. Gentsch T.O., Larsen P.B., Daughtry D.C., Chesney J.G., Spear H.C. Community-wide availability of pulmonary embolectomy with cardiopulmonary bypass. Ann Thorac Surg 1969;7:97-103.[Free Full Text]
14. Robiosek F., Tam W., Daugherty H.K., Mullen D.C. Myocardial protection during open-heart surgery. Coronary perfusion versus topical cardiac hypothermia. Ann Thorac Surg 1970;10:340-353.[Free Full Text]
15. Gonzalez-Lavin L., Al-Janabi N., Ross D.N. Long-term results after aortic valve replacement with preserved aortic homografts. Ann Thorac Surg 1972;13:594-606.[Abstract/Free Full Text]
16. Stoney W.S., Alford W.C., Burrus G.R., Thomas C.S. Repair of anteroseptal ventricular aneurysm. Ann Thorac Surg 1973;15:394-404.[Abstract/Free Full Text]
17. Trinkle J.K., Marcos J., Grover F.L., Cuello L.M. Management of the wounded heart. Ann Thorac Surg 1974;17:230-236.[Abstract/Free Full Text]
18. Bender H.W., Fisher R.D., Faulkner S.L., Friesinger G.C. Unstable coronary artery disease. Ann Thorac Surg 1975;19:521-528.[Abstract/Free Full Text]
19. Parker F.B., Neville J.F., Hanson E.L., Mohiuddin S., Webb W.R. Management of acute aortic dissection. Ann Thorac Surg 1975;19:436-442.[Abstract/Free Full Text]
20. Donahoo J.S., Brawley R.K., Gott V.L. The heparin-coated vascular shunt for thoracic aortic and great vessel procedures: a ten-year experience. Ann Thorac Surg 1977;23:507-513.[Abstract/Free Full Text]
21. West P.N., Ferguson T.B., Clark R.E., Weldon C.S. Multiple valve replacement: changing status. Ann Thorac Surg 1978;26:32-37.[Abstract/Free Full Text]
22. Craver J.M., Jones E.L., McKeown P., Bone D.K., Hatcher C.R., Kandrach M. Porcine cardiac xenograft valves: analysis of survival, valve failure, and explanation. Ann Thorac Surg 1982;34:16-21.[Abstract/Free Full Text]
23. Barner H.B., Swartz M.T., Mudd J.G., Tyras D.H. Late patency of the internal mammary artery as a coronary bypass conduit. Ann Thorac Surg 1982;34:408-412.[Abstract/Free Full Text]
24. Urschel H.C., Razzuk M.A., Leshnower A.C. Bypass grafting and aneurysmorrhaphy for aortic arch aneurysms. Ann Thorac Surg 1983;35:579-583.[Abstract/Free Full Text]
25. Livesay J.J., Cooley D.A., Reul G.J., et al. Resection of aortic arch aneurysms: a comparison of hypothermic techniques in 60 patients. Ann Thorac Surg 1983;36:19-28.[Abstract/Free Full Text]
26. Tavares S., Hankins J.R., Moulton A.L., et al. Management of penetrating cardiac injuries: the role emergency room thoracotomy. Ann Thorac Surgery 1984;38:183-187.[Abstract/Free Full Text]
27. Myers W.O., Davis K., Foster E.D., Maynard C., Kaiser G.C. Surgical survival in the coronary artery surgery study (CASS) registry. Ann Thorac Surg 1985;40:245-260.[Abstract/Free Full Text]
28. Frist W.H., Oyer P.E., Baldwin J.C., Stinson E.B., Shumway N.E. HLA compatibility and cardiac transplant recipient survival. Ann Thorac Surg 1987;44:242-246.[Abstract/Free Full Text]
29. Kanter K.R., McBride L.R., Pennington D.G., et al. Bridging to cardiac transplantation with pulsatile ventricular assist devices. Ann Thorac Surg 1988;46:134-140.[Abstract/Free Full Text]
30. Vauthey J.-N., Berry D.W., Snyder D.W., et al. Left ventricular aneurysm repair with myocardial revascularization: an analysis of 245 consecutive patients over 15 years. Ann Thorac Surg 1988;46:29-35.[Abstract/Free Full Text]
31. Greene P.S., Cameron D.E., Augustine S., Gardner T.J., Reitz B.A., Baumgartner W.A. Exploratory analysis of time-dependent risk for infection rejection and death after cardiac transplantation. Ann Thorac Surg 1989;47:650-654.[Abstract/Free Full Text]
32. Galloway A.C., Colvin S.B., Baumann F.G., et al. A comparison of mitral valve reconstruction with mitral valve replacement: intermediate-term results. Ann Thorac Surg 1989;47:655-662.[Abstract/Free Full Text]
33. Edwards F.H., Albus R.A., Zajtchuk R., Graeber G.M., Barry M. A quality assurance model of operative mortality in coronary artery surgery. Ann Thorac Surg 1989;47:646-649.[Abstract/Free Full Text]
34. Arom K.V., Nicoloff D.M., Kersten T.E., Northrup W.F., III, Lindsay W.G., Emery R.W. Ten years' experience with the St. Jude Medical valve prosthesis. Ann Thorac Surg 1989;47:831-837.[Abstract/Free Full Text]
35. Gott V.L., Pyeritz R.E., Cameron D.E., Greene P.S., McKusick V.A. Composite graft repair of Marfan aneurysm of the ascending aorta: results in 100 patients. Ann Thorac Surg 1991;52:38-45.[Abstract/Free Full Text]
36. Pate J.W., Cole F.H., Jr, Walker W.A., Fabian T.C. Penetrating injuries of the aortic arch and its branches. Ann Thorac Surg 1993;55:586-592.[Abstract/Free Full Text]
37. Safi H.J., Brien H.W., Winter J.N., et al. Brain protection via cerebral retrograde perfusion during aortic arch aneurysm repair. Ann Thorac Surg 1993;56:270-276.[Abstract/Free Full Text]
38. Jones D.R., Hill R.C., Vasilakis A., et al. Safe use of heparin-coated bypass circuits incorporating a pump-oxygenator. Ann Thorac Surg 1994;57:815-819.[Abstract/Free Full Text]
39. Craver J.M., Bufkin B.L., Weintraub W.S., Guyton R.A. Neurologic events after coronary bypass grafting: further observations with warm cardioplegia. Ann Thorac Surg 1995;59:1429-1433.[Abstract/Free Full Text]
40. Langenburg S.E., Buchanan S.A., Blackbourne L.H., et al. Predicting survival after coronary revascularization for ischemic cardiomyopathy. Ann Thorac Surg 1995;60:1193-1196.[Abstract/Free Full Text]
41. Hammon J.W., Stump D.A., Kon N.D., et al. Risk factors and solutions for the development of neurobehavioral changes after coronary artery bypass grafting. Ann Thorac Surg 1997;63:1613-1617.[Abstract/Free Full Text]
42. Niwaya K., Knott-Craig C.J., Santangelo K., Lane M.M., Chandrasekaran K., Elkins R.C. Advantage of autograft and homograft valve replacement for complex aortic valve endocarditis. Ann Thorac Surg 1999;67:1603-1608.[Abstract/Free Full Text]
43. Franga D.L., Kratz J.M., Crumbley A.J., Zellner J.L., Stroud M.R., Crawford F.A. Early and long-term results of coronary artery bypass grafting in dialysis patients. Ann Thorac Surg 2000;70:813-818.[Abstract/Free Full Text]
44. Knott-Craig CJ, Elkins RC, Santangelo K, McCue C, Lane MM. Aortic valve replacement: comparison of late survival between autografts and homografts. Ann Thorac Surg 2000;69:1327–31
45. Merrill W.H., Friesinger G.C., Graham T.P., Jr, et al. Long-lasting improvement after septal myectomy for hypertrophic obstructive cardiomyopathy. Ann Thorac Surg 2000;69:1732-1735.[Abstract/Free Full Text]
46. John R, Rajasinghe HA, Chen JM, et al. Long-term outcomes after cardiac transplantation: an experience of immunosuppressive therapy. Ann Thorac Surg 2001;72:440–9
47. Cleveland J.C., Shroyer L.W., Chen A.Y., Peterson E., Grover F.L. Off-pump coronary artery bypass grafting decreases risk-adjusted mortality and morbidity. Ann Thorac Surg 2001;72:1282-1289.[Abstract/Free Full Text]
48. Magee M.J., Dewey T.M., Acuff T., et al. Influence of diabetes on mortality and morbidity: off-pump coronary artery bypass grafting versus coronary artery bypass grafting with cardiopulmonary bypass. Ann Thorac Surg 2001;72:3.[Free Full Text]
49. Kon N.D., Riley R.D., Adair S.M., Kitzman D.W., Cordell A.R. Eight-year results of aortic root replacement with the freestyle stentless porcine aortic root bioprosthesis. Ann Thorac Surg 2002;73:1817-1821.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Arom, K. V. Articles by Grover, F. L. Search for Related Content PubMed PubMed Citation Articles by Arom, K. V. Articles by Grover, F. L. Related Collections History