14. De Leval MR, McGoon DC, Wallace RB, Danielson GK, Mair DD: Management of truncal valvular regurgitation. Ann Surg 180: 427, 1974 15. Mair DD, Ritter DG, Davis GD, Wallace RB, Danielson GK, McGoon DC: Selection of patients with truncus arteriosus for surgical correction: Anatomic and hemodynamic considerations. Circulation 49: 144, 1974 16. Arai T, Tsuzuki Y, Nogi M, Kurashige K, Koyanagi H, Nishida H, Ikeda Y, Ichikawa H: Experimental study on bypass between the right ventricle and pulmonary artery, left ventricle and pulmonary artery, and left ventricle and aorta by means of homograft with valve. Bull Heart Inst Japan 9: 49, 1965 17. Rastelli GC, Titus JL, McGoon DC: Homograft of ascending aorta and aortic valve as a right ventricular outflow: An experimental approach to the repair of truncus arteriosus. Arch Surg 95: 698, 1967 18. DuShane JW: Cited by DuShane JW, Kirklin JW: Late results of the repair of ventricular septal defect on pulmonary vascular disease. In Advances in Cardiovascular Surgery. Editedby JW Kirklin. New York, Grune & Stratton, 1973, pp 9-16 19. Parker RK, McGoon DC, Danielson GK, Wallace RB, and Mair DD: Repair of truncus arteriosus in patients with prior banding of the pulmonary artery. Surgery 78: 761, 1975 20. Gelband H, Van Meter S, Gersony WM: Truncal valve abnormalities in infants with persistent truncus arteriosus: A clinicopathologic study. Circulation 45: 397, 1972 21. McGoon DC, Wallace RB, Danielson GK: Homografts in reconstruction of congenital cardiac anomalies: Expanded operability in complex congenital heart disease. Mayo Clin Proc 47: 101, 1972 22. McGoon DC, Wallace RB, Danielson GK: The Rastelli operation: Its indications and results. J Thorac Cardiovasc Surg 65: 65, 1973 23. Brawley RK, Gardner TJ, Donahoo JS, Neill CA, Rowe RD, Gott VL: Late results after right ventricular outflow tract reconstruction with aortic root homografts. J Thorac Cardiovasc Surg 64: 314, 1972 24. Kaplan S, McKinivan CE, Helmsworth JA, Benzing G III, Schwartz DC, Schreiber JT: Complications following homograft replacement of the right ventricular outflow tract. Ann Thorac Surg 18: 250, 1974 25. Merin G, McGoon DC: Reoperation after insertion of aortic homograft

14. De Leval MR, McGoon DC, Wallace RB, Danielson GK, Mair DD: Management of truncal valvular regurgitation. Ann Surg 180: 427, 1974 15. Mair DD, Ritter DG, Davis GD, Wallace RB, Danielson GK, McGoon DC: Selection of patients with truncus arteriosus for surgical correction: Anatomic and hemodynamic considerations. Circulation 49: 144, 1974 16. Arai T, Tsuzuki Y, Nogi M, Kurashige K, Koyanagi H, Nishida H, Ikeda Y, Ichikawa H: Experimental study on bypass between the right ventricle and pulmonary artery, left ventricle and pulmonary artery, and left ventricle and aorta by means of homograft with valve. Bull Heart Inst Japan 9: 49, 1965 17. Rastelli GC, Titus JL, McGoon DC: Homograft of ascending aorta and aortic valve as a right ventricular outflow: An experimental approach to the repair of truncus arteriosus. Arch Surg 95: 698, 1967 18. DuShane JW: Cited by DuShane JW, Kirklin JW: Late results of the repair of ventricular septal defect on pulmonary vascular disease. In Advances in Cardiovascular Surgery. Editedby JW Kirklin. New York, Grune & Stratton, 1973, pp 9-16 19. Parker RK, McGoon DC, Danielson GK, Wallace RB, and Mair DD: Repair of truncus arteriosus in patients with prior banding of the pulmonary artery. Surgery 78: 761, 1975 20. Gelband H, Van Meter S, Gersony WM: Truncal valve abnormalities in infants with persistent truncus arteriosus: A clinicopathologic study. Circulation 45: 397, 1972 21. McGoon DC, Wallace RB, Danielson GK: Homografts in reconstruction of congenital cardiac anomalies: Expanded operability in complex congenital heart disease. Mayo Clin Proc 47: 101, 1972 22. McGoon DC, Wallace RB, Danielson GK: The Rastelli operation: Its indications and results. J Thorac Cardiovasc Surg 65: 65, 1973 23. Brawley RK, Gardner TJ, Donahoo JS, Neill CA, Rowe RD, Gott VL: Late results after right ventricular outflow tract reconstruction with aortic root homografts. J Thorac Cardiovasc Surg 64: 314, 1972 24. Kaplan S, McKinivan CE, Helmsworth JA, Benzing G III, Schwartz DC, Schreiber JT: Complications following homograft replacement of the right ventricular outflow tract. Ann Thorac Surg 18: 250, 1974 25. Merin G, McGoon DC: Reoperation after insertion of aortic homograft

Between 1965 and 1972, 250 patients (186 men and 64 women; median age, 48 years) underwent aortic valve replacement with homografts preserved with beta-propiolactone (98 patients) or irradiation (152 patients); operative mortality was 6% (15 patients). Follow-up to death, reoperation, or recent evaluation was completed in 95% of patients, and median follow-up time was 11.4 years. Thromboembolic events occurred in two patients (0.21 events/100 patients/yr). Risk of reoperation was 22% at 5 years, 62% at 10 years, 85% at 15 years, and 95% at 20 years. Factors associated with increased risk of reoperation were young age, male sex, native aortic valve insufficiency, previous aortic valve surgery, history of endocarditis, and larger homograft size. Survival at 5, 10, 15, and 20 years was 85%, 66%, 53%, and 38%, respectively. Late survival was diminished in older patients and those with native aortic valve insufficiency and coronary artery disease. Aortic valve replacement with nonviable homografts has a high risk of late tissue degeneration and need for reoperation. However, the incidence of thromboemboli is minimal, and grafts can be replaced with low risk (4.5%) so that overall patient survival is similar to that observed with other bioprosthetic or mechanical heart valves.

Antiplatelet therapy with dipyridamole, 100 mg q.i.d., starting 2 days before surgery, followed by aspirin, 325 mg t.i.d. plus dipyridamole, 75 mg t.i.d., 7 hours after surgery was assessed in the prevention of saphenous vein bypass graft occlusion. Early (less than or equal to 1 month) and late (1 year) occlusions were reduced both on a per patient and a per distal anastomosis basis. Bleeding complications were not increased. Graft occlusion in high-risk situations (low-flow grafts and endarterectomy) was reduced, but not eliminated, by this antiplatelet regimen. The authors recommend this combination of dipyridamole before surgery, adding aspirin after surgery, to prevent coronary artery bypass graft occlusion.

Despite the use of oral anticoagulation in patients with prosthetic heart valves, persistent thromboembolism over time warrants a search for improved methods of prevention. Thus, patients receiving 1 or more mechanical prosthetic heart valves were randomized to therapy with warfarin plus dipyridamole (400 mg/day) or warfarin plus aspirin (500 mg/day) on the basis of location and type of valve and surgeon, and followed up with a concurrent, nonrandomized control group taking warfarin alone. In 534 patients followed up 1,319 patient-years, excessive bleeding (necessitating blood transfusion or hospitalization) was noted in the warfarin plus aspirin group (23 of 170 [14%], or 6.0/100 patient-years) compared with warfarin plus dipyridamole (7 of 181 [4%], or 1.6/100 patient-years, p less than 0.001), or warfarin alone (9 of 183 [5%], or 1.8/100 patient-years, p less than 0.001). A trend was evident toward a reduction in thromboembolism in the warfarin plus dipyridamole group (2 of 181 [1%], or 0.5/100 patient-years) as compared with warfarin plus aspirin (7 of 170 [4%], or 1.8/100 patient-years), or warfarin alone (6 of 183 [4%], or 1.2/100 patient-years). Adequacy of anticoagulation (based on 12,720 prothrombin time determinations) was similar in all 3 groups with 65% of prothrombin times in the therapeutic range (1.5 less than or equal to prothrombin time/control less than or equal to 2.5), 30% too low, and 5% too high. Warfarin plus aspirin therapy resulted in excessive bleeding and is contraindicated. Longer follow-up study is needed to determine whether further separation of the incidence of thromboembolism can be detected.

To prevent occlusion of aortocoronary-artery-bypass grafts, we conducted a prospective, randomized-double-blind trial comparing dipyridamole (instituted two days before operation) plus aspirin (added seven hours after operation) with placebo in 407 patients. Vein-graft angiography was performed in 360 patients (88 per cent) within six months of operation (median, eight days). Within one month of operation, 3 per cent of vein-graft distal anastomoses (10 of 351) were occluded in the treated patients, and 10 per cent (38 of 362) in the placebo group; the proportion of patients with one or more distal anastomoses occluded was 8 per cent (10 of 130) in the treated group and 21 per cent (27 of 130) in th placebo group. This benefit in graft patency persisted in each of over 50 subgroups. Early postoperative bleeding was similar in the two groups. In this trial dipyridamole and aspirin were effective in preventing graft occlusion early after operation.

The current status of the septation procedure for univentricular heart was assessed by analyzing the outcomes of 45 procedures done from 1973 through June of 1978. The hospital mortality was 21 (47%) of 45 patients, and eight (18%) patients died during follow-up. Of the 16 survivors, 12 have a fair-to-good surgical result and four are doing poorly. Only two factors were significantly related to increased survival: no history of congestive failure and no previous palliative procedure. Multivariate analysis revealed a small subgroup of 11 patients with a survival rate of 82%. These patients had a subaortic chamber to the left, no previous congestive failure, no previous palliation, and no profound polycythemia, and they were between 5 and 15 years of age at operation. Current practice now employs the Fontan procedure for patients with univentricular heart and low pulmonary artery pressure and resistance. However, nearly half of the patients with univentricular heart will not qualify for a Fontan procedure. For a selected group of these patients, septation must be considered because of the poor long-term outlook for patients with univentricular heart who do not undergo operation.

The entire 21-year experience at our institution with intracardiac repair of corrected transposition of the great arteries (CTGA) associated with ventricular septal defect (VSD) is reviewed. There were 53 such operations. The hospital mortality rate fell from 50% before 1972 to 18% since then (p < 0.02), and was 19% when a systemic ventriculotomy was not used compared to 60% when it was used (p < 0.01). A technique is described for correction of CTGA with VSD and pulmonary outflow tract obstruction (POTO) designed to minimize risk of heart block and to increase the degree of relief of POTO: the VSD patch is placed to the right side of the conduction tissue and the pulmonary outflow tract, and POTO is bypassed with an extracardiac conduit. This technique reduced the incidence of complete heart block from 67% by direct relief of POTO to 28% (p < 0.05) and decreased the frequency of inadequate relief of POTO from 50% to 8% (p < 0.01). However, experience with these two approaches was not concurrent. Successful intraoperative mapping of the conduction bundle was associated with operatively induced heart block in 37%, whereas prior to the availability of mapping the incidence of heart block was 57% (difference not significant). Twelve patients required atrioventricular (AV) valve repair or replacement, and in them the operative mortality was 50%, compared to 20% in the remainder (p < 0.05). For the 21 patients with VSD and POTO who achieved good relief of POTO, the operative mortality was only 5%. These findings suggest that good progress has been made in the surgical repair of CTGA with VSD, though further improvement in results is required.

The presence of hypoplastic pulmonary arteries in the symptomatic patient with pulmonary atresia and ventricular septal defect (VSD) precludes one-stage complete correction and mandates some type of preliminary palliative procedure. Since 1976, we have palliated 38 patients by establishment of right ventricular-pulmonary arterial continuity (right ventricular outflow construction) without closure of the VSD. The ages ranged from 2 to 54 years (mean 13.3); 12 patients had 15 prior systemic-pulmonary arterial shunts. There were three hospital deaths (8%), all in patients with prior shunts; no deaths occurred in the 26 patients without shunts (p = 0.05). All survivors manifested clinical improvement; there was a mean decrease in hemoglobin concentration of 3.6 gm/dl (p < 0.001) and a mean increase in arterial oxygen saturation of 15.2% (p < 0.001). Cardiac catheterization, performed in 12 patients 10 to 26 months (mean 17.2) after operation, revealed a 2.20-fold increase in mean pulmonary artery size (p = 0.0001), defined as the ratio of the sum of the diameters of the right and left pulmonary arteries to the diameter of the descending aorta. Three patients have now undergone completion of correction by placement of a valved conduit from right ventricle to pulmonary artery with patch closure of the VSD. These results were compared with those obtained in a comparable group of 13 patients with pulmonary atresia, VSD, and similarly hypoplastic pulmonary arteries treated initially with systemic-pulmonary arterial shunting. The mean pulmonary arterial enlargement of only 1.55 times initial size detected on late restudy (mean 33.4 months postoperatively), p = 0.005, was statistically less than that of the reconstruction group (p < 0.05), was less predictable, and was associated with more pulmonary arterial and anastomotic distortion (46% versus 18%). The results of right ventricular outflow construction suggest that this is a more satisfactory initial procedure for the management of the symptomatic patient over 2 years of age who has pulmonary atresia, VSD, and hypoplastic pulmonary arteries.

Although ultrasonic and computed tomographic technology is continuing to evolve, these already sophisticated imaging tests have become the primary methods of evaluating the patients with aortic aneurysms. Both modalities are accurate and noninvasive. The major value of these techniques to the surgeon is in assessing those patients in whom the diagnosis is in doubt and in following patients in whom it has been elected to observe a small stable aneurysm. Changes in an aneurysm can be detected on serial examinations which might indicate the need for surgery in spite of adverse patient risk factors, and these techniques may provide a means of detecting false aneurysms in patients who are difficult to assess accurately by physical examination. Ultrasound and computed tomography are very helpful techniques in the diagnosis and assessment of aortic aneurysms. These diagnostic modalities are not required in every patient and should be used discriminately and when indicated.

The coronary arterial anatomy in 26 univentricular hearts, its relation to the morphologic characteristics of the ventricles and rudimentary chambers, and its surgical implications were analyzed. All of the hearts except two had been operated on; 18 had septation with or without an extracardiac conduit and 6 had had palliative procedures. Twenty-one univentricular hearts with a left ventricular type main chamber had an anterior outlet chamber (17 left-sided subaortic, 3 right-sided subaortic and 1 right-sided subpulmonary). Right and left delimiting arteries outlined the outlet chamber in 16 hearts (76 percent). In 20 of the 21 hearts, large delimiting parallel branches of the right coronary artery coursed over the anterior wall of the heart; 13 of these vessels had been injured surgically with resultant ischemic myocardial necrosis. Five univentricular hearts did not have an outlet chamber; two had a left ventricular type main chamber and three had a morphologically right ventricular main chamber. Three of the five hearts had rudimentary pouches, located anteriorly in one and posteriorly in two. The two rudimentary pouches lying posteriorly were not outlined by delimiting arteries. Two of the five univentricular hearts without an outlet chamber also had injured coronary arteries. Thus, the identification of outlet chambers and rudimentary pouches in univentricular hearts is facilitated by the determination of coronary anatomy. The presence of major delimiting parallel branches over the usually favored ventriculotomy sites renders them vulnerable to surgical injury; such mishaps occurred in 15 of the 24 hearts that had either corrective or palliative operations.

Corrective operations were attempted in 10 patients with straddling atrioventricular valve (SAVV), and successful palliative operations were performed in another nine patients. SAVV occurred as a part of five different cardiac complexes, and three anatomic types were encountered. The SAVV was a tricuspid valve in eight patients, right atrioventricular valve (in univentricular hearts) in five, and mitral valve in six. Four types of corrective procedures were used: (1) the Rastelli operation, (2) isolated closure of the universally present ventricular septal defect by a patch deviating around the straddling portion of the SAVV, (3) replacement of the SAVV with or without insertion of an extracardiac conduit, and (4) closure of the SAVV plus right atrium-pulmonary artery anastomosis (modified Fontan procedure). All three hospital deaths occurred in patients managed by replacement of SAVV and extracardiac conduit insertion (technique 3). This experience allows preliminary speculation about preferred methods of surgical management.

A review is made of all Mayo Clinic cases wherein 468 patients have received 516 extracardiac conduits in the repair of congenital heart defects. All patients had complex defects, which are classified in 10 basic diagnostic categories. The early mortality rate (which ranged from 4% to 49%, according to diagnostic group) averaged 25%, improving with experience. The postoperative complication rate was 70%. In 1% of operations, compression of the conduit was encountered at chest closure. The conduits placed in 333 patients contained a porcine valve. The conduit diameter averaged 22 mm. At completion of the operation, the mean gradient across the conduit was 22.8 mm Hg. At late study this mean gradient remained at 26 mm Hg in patients who had received the currently employed porcine-valved conduit. The average ratio of intraoperative postrepair ventricular (RV/LV) systolic pressures was 0.68. It was lower among the group who were to survive but was not ov prognostic value in individual cases. Among patients who survived the postoperative period, the late mortality rate (based on a mean of 3.6 years' follow-up) averaged 3.5% per year. Of the long-term survivors, 18% have required reoperation, primarily for replacement of calcified aortic homograft conduits used in the earlier part of the experience. Only 0.8% of porcine-valved conduits have required reoperation. Unrestricted life-styles were possible for 90% of survivors.