after Repair of Tetralogy of Fallot

Transcription

1 Chronic Congestive Heart Failure after Repair of Tetralogy of Fallot ALBERT P. ROCCHINI, M.D., AMNON ROSENTHAL, M.D., MICHAEL FREED, M.D., ALDO R. CASTANEDA, M.D., AND ALEXANDER S. NADAS, M.D. SUMMARY The etiology of chronic congestive heart failure (CHF) after repair of tetralogy of Fallot was determined in 12 patients consecutively catheterized 1 to 12 years postoperatively. Chronic CHF was observed in 36/12 patients. The most prevalent abnormality leading to congestive failure (31/36) was a large residual ventricular septal defect alone or in combination with other lesions. All postoperative patients with pulmonary to systemic flow ratios greater than 2:1 (25/12) had congestive failure and evidence of biventricular dysfunction. Significant tricuspid regurgitation (N = 11) and persistent systemic to pulmonary artery shunts (N = 6) contributed to volume overload and congestive failure in the patients with large residual ventricular septal defect. Isolated severe residual right ventricular outflow tract obstruction was an uncommon cause of chronic CHF. Pulmonary artery hypertension was present in 2/36 patients with CHF. The increased pulmonary pressure was not wholly due to an increased pulmonary flow since 7/2 patients had pulmonary vascular resistance greater than 3 mm Hg/L/min/m2. Our findings indicate that persistent or chronic congestive heart failure in postoperative tetralogy of Fallot patients requires bilateral cardiac catheterization since an identifiable and surgically correctable lesion is nearly always present. Downloaded from by on October 17, 218 THE MAJORITY OF PATIENTS WITH TETRALOGY OF FALLOT (TF) are asymptomatic after intracardiac repair, but continue to have mild residual auscultatory, radiographic or electrocardiographic abnormalities.1-' In some asymptomatic patients, cardiac catheterization may reveal persistent severe defects unsuspected on clinical examination.6 1 However, there are a small number of patients who do not improve after repair of TF and, in fact, some of these patients may even become more symptomatic. Most patients with persistent symptomatology have chronic congestive heart failure (CHF).7 It is this group of patients with chronic congestive heart failure that forms the basis for our study. Through review of the clinical and hemodynamic data on 12 patients consecutively catheterized one to 12 years (mean 2.3 years) after surgical repair of TF, we plan to document the etiological factors that may result in chronic congestive heart failure after surgical repair. Material and Methods From January 1965 to June 1975, 12 patients underwent cardiac catheterization at The Children's Hospital Medical Center of Boston, one to 12 years (mean 2.3 years), following repair of TF. The records, hemodynamic data, and angiograms of the patients were reviewed. The repair of TF was performed at.8 to 28 years of age (mean 1.7) at either The Children's Hospital Medical Center of Boston (98/12), the Mayo Clinic (2/12) or University of Alabama (2/12). Twenty of 12 patients were catheterized prior to Prior to 1972, postoperative cardiac catheterization at our institution was performed only when the patient was symptomatic. Since 1972, however, cardiac catheterization has been recommended to all patients, one year after repair. Six- From the Departments of Cardiology and Cardiovascular Surgery, The Children's Hospital Medical Center and the Departments of Pediatrics and Surgery, Harvard Medical School, Boston, Massachusetts. Supported in part by Training Grant #H15855 and Program Project Grant #HL136 from the National Institutes of Health. Address for reprints: Albert P. Rocchini, M.D., Department of Cardiology, The Children's Hospital Medical Center, 3 Longwood Avenue, Boston, Massachusetts Received February 7, 1977; revision accepted March 2, ty percent of the patients operated on between 1972 and 197 had postoperative catheterization. A previous systemic-to-pulmonary artery anastomosis had been performed in 77/12 patients: 16 patients had a Potts shunt alone or in combination with a Blalock-Taussig shunt, 15 patients a Waterston shunt alone or in combination with a Blalock-Taussig shunt and 6 patients had a Blalock-Taussig shunt. The right ventricular outflow tract was reconstructed by infundibulectomy with or without pulmonary valvotomy in 37/12, right ventricular outflow patch extending to below the pulmonary anulus in 32/12 patients (pericardial in 21/32 and prosthetic material in 11/23 patients); and right ventricular outflow patch extending across the pulmonary anulus in 2/12 (pericardial in 12/2, prosthetic material in 12/2 patients). A valved conduit was interposed between the right ventricular outflow tract and pulmonary artery in 9/12 patients (heterograft as provided by Hancock Laboratories in six patients and aortic homograft in three) (table 1). The ventricular septal defect was closed with an Ivalon patch in /12 patients and a Dacron patch in 58/12 patients. All 12 patients underwent right heart catheterization and 9/12 had right ventricular angiography performed. Left heart catheterization was performed in 86/12 patients, left ventricular angiography in 6 patients, and ascending aortic angiography in 2 of the patients. The following hemodynamic measurements were made: a) right atrial, right ventricular and aortic and/or femoral artery pressures and oxygen saturation were measured in all 12 patients; b) pulmonary artery pressure and oxygen saturation was measured in 99/12 patients; c) left ventricular pressures were measured in 86/12 patients; d) cardiac output was determined by the Fick principle using a measured oxygen consumption in 93/12 patients and an assumed oxygen consumption value' in nine patients. A left-to-right shunt at ventricular level was confirmed by oxygen saturation measurements and/or left ventricular angiography and its site localized by left ventricular angiography and/or at surgery. Pulmonary insufficiency was graded by a numerical scale. Regurgitation was considered mild if a grade 1-2 short, early diastolic murmur was present, moderate if a grade 3 or greater murmur was audible

2 36 CIRCULATION VOL 56, No 2, AUGUST 1977 Downloaded from by on October 17, 218 TABLE 1. Congestive Heart Failure (CHF) Related to the Type of Surgical Procedure and Presence of Residual Pulmonary Stenosis (PS) Total CHF PS (> mm Hg) No. No. Total CHF No right ventricular outflow patch Right ventricular outflow patch to the pulmonary anulus Right ventricular outflow patch across the pulmonary anulus 2 8 Right ventricle to pulmonary artery conduit Totals and severe if a loud murmur was associated with equal pulmonary diastolic and right ventricular end-diastolic pressures. Tricuspid regurgitation was identified by right ventricular angiography, a loud regurgitant murmur with inspiratory augmentation, and a right atrial pressure tracing with a prominent, broad V wave and near or complete obliteration of the x descent and a steep y descent. Chronic CHF was documented by presence of marked cardiomegaly, hepatomegaly, and the clinical dependence on digitalis and diuretics. Results Chronic CHF one year or more after surgical repair was observed in 36/12 patients with TF. Table 2 lists the clinical findings and multiple hemodynamic residua and sequelae observed in these 36 patients. The most prevalent abnormality leading to chronic congestive heart failure was a residual large ventricular septal defect, alone or in combination with other lesions. Residual Ventricular Septal Defect A residual ventricular septal defect (VSD) was identified in 58/12 patients; among the 58 patients with a ventricular defect, 31 had chronic congestive heart failure. A pulmonary-to-systemic flow ratio (Qp/Qs) - 2 was present in 28/31 patients with a residual VSD and CHF. The other three patients had a large residual VSD but a small left-toright shunt because of marked right ventricular outflow tract obstruction. The relationship between the magnitude of the left-to-right shunt via the VSD and the presence of CHF is shown in figure 1. All postoperative patients with a Qp/Qs > 2 had chronic CHF. The residual VSD was localized solely to the region of the VSD patch in 28/31 patients with CHF. In two patients, an additional large muscular VSD, unrecognized prior to and during surgical repair, was present. The third patient (table 2; no. 2) developed acute symptoms of CHF five years after surgical repair of TF. At reoperation, a calcified and fractured Ivalon patch was found. Residual Pulmonary Stenosis Significant residual right ventricular outflow obstruction NUMBER of PATIENTS N-58 m CHF < > 2 C.WVphy.Iy) FIGURE 1. The effect ofpulmonary-to-systemic flow ratio on the development of congestive heart failure (CHF) in 58 patients with a residual ventricular septal defect after repair of tetralogy of Fallot. Note that allpatients with aflow ratio ofgreater than 2 have CHF. (pressure gradient ' mm Hg) was observed in 29/12 patients, and 16/29 had CHF (table 1). However, 1 of the 16 patients with significant outflow obstruction also had large residual VSD. The two patients with congestive heart failure, right ventricular outflow obstruction, and no residual VSD were patients 31 and 36 (table 2). In patient 36, the right ventricular outflow patch was not extended across the pulmonary valve anulus at surgical repair. Postoperative catheterization demonstrated suprasystemic right ventricular pressure, a large right ventricular outflow aneurysm, and a hypoplastic pulmonary valve anulus. At reoperation, the outflow tract was reconstructed with a transanular patch. Postoperatively, CHF disappeared and a subsequent cardiac catheterization, one year later demonstrated normal right ventricular pressures (table 2). Surgical repair in patient 31 involved the use of an aortic homograft. Postoperative catheterization demonstrated severe tricuspid regurgitation and marked elevation of right ventricular end-diastolic pressure in addition to right ventricular tract obstruction. Residual Aortopulmonary Shunts Twenty-one of 12 patients had residual aortopulmonary shunts: a Blalock-Taussig anastomosis in ten, Potts anastomosis in six, ascending aorta to pulmonary artery shunt in four, and a large aortopulmonary collateral vessel in one. Congestive heart failure was present in 9/21 patients with residual aortopulmonary shunts and all but one of these patients had additional hemodynamically significant residual lesions (table 2). Most of the residual aortopulmonary shunts were small. An attempt had been made to close them at the initial operation but either was incomplete or the Blalock-Taussig shunt recanalized. Left Blalock- Taussig shunts were particularly difficult to close. The exceptional patient (table 2; no. 33) had a large left-to-right shunt and congestive heart failure secondary to a big aortopulmonary collateral vessel from the descending aorta to right pulmonary artery. Ligation of the aortopulmonary collateral vessel resolved the CHF.

3 ti l l. < _ M X > N - n + co X + mp _ n O t t X -~~~ CHRONIC CHF IN POSTOP TFIRocchini et al. 37 S S i z 9 ; 9; \ax,9t~~~~~~~~~~~~~~~~-go X B.n-w~~~~~~~~ '- ;ll < H H H H H H H H cs H H H,, H H ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Hvev~~~~, k.s s O 5 + H X e H O H O O b >X e e XXs<>O>o<>O bex g d +m I l+l ~~~~~~~~~~~~~~~~~~~~~~~~> > ~~~~~~~~~~~~~~~~~~~~~~~~ : l: : :<<>> n<oo ::b<tt>s Q- A o Oe< oc8cs8 cz.oo wcog< e + g = EmE A~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ a : n. so :+oo s1221 t cz e P e~~~~~~~~~~~~~~~~~~~~~~~~ Downloaded from by on October 17, 218 C cl cc6 ci cl C'scl ci t C6 cl C6 Ci (, ci C- (. C6-C6 CimCi> ' ic m Or.1 Cr co Co t OCtcoCQ CD ' It LO t Cq t XC9 1 1 CCo ec etc C n? 9eOC A9C )a9+oon t-oocqcoc <<>L : Ac 1t t o sxtq -> Cbbb<O > >*ego CD sw vs~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. va;-~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~U' C)CI~1t qc LOA Iq Eo C) C CS td M co t- co co i>a to ko t- Lo m to co _- f, Lo to tnbc -P CD LO to LO O o C) I I. Io Lo Lo co Ak~L oc L tt C- ~~~~~~~~~~~~~~~~~~2d C t I OOXCO _ OC NNCO - C. --N -N- NNNN- o- Q,, J,3 _1 NNCoCtCDt-- ) 5 --No_C A.c <QbXo o oo m q m co U X o _It 'It C) C) 1 Lo Lo N co _~m b z HN ~,,M 9 NCSSSS9lSXO ABPq-ir-- -1-

4 38 CIRCULATION VOL 56, No 2, AUGUST 1977 Downloaded from by on October 17, 218 Pulmonary Regurgitation A murmur of pulmonary insufficiency was identified in 88/12 patients, while only 28/88 patients had moderate to severe pulmonary insufficiency. None of these 28 patients had a right ventricular to pulmonary artery gradient > 5 mm Hg and in no patient was pulmonary regurgitation the only hemodynamic lesion responsible for CHF. The degree of pulmonary insufficiency was not significantly different between patients with and without pulmonary artery hypertension (table 2). Tricuspid Regurgitation Moderate to severe tricuspid regurgitation was observed in 11/12 patients. All 11 patients had chronic CHF, ten of the 11 had an associated large left to right shunt via a VSD and five of the 11 had moderate to severe pulmonary regurgitation. Severe right ventricular obstruction or pulmonary vascular obstruction was present in seven of the 11 patients. All of these patients had both extensive right ventricular muscle resected and had their VSD patch anchored to a portion of the tricuspid valve during their surgical repair. Pulmonary Artery Hypertension A pulmonary artery systolic pressure ' 35 mm Hg was measured in 23/12 patients. Among these 23 patients with increased pulmonary artery pressure, CHF was present in 2, a large residual VSD in 19, and a residual systemic to pulmonary shunt in 8. A large left-to-right shunt via a VSD and/or shunt was the most important factor contributing to the increased pulmonary artery pressure in postoperative patients (fig. 2). Preoperative pulmonary artery pressure measured in 17/23 patients was elevated in seven patients. All these patients had a previous Potts or Waterston shunt. No patient with a preoperative pulmonary artery pressure of greater than 35 mm Hg and a residual VSD or aortopulmonary shunt had a normal pulmonary artery pressure following surgical repair. The increase in pulmonary artery pressure was not wholly I z 9 gx c 2 f * o * o i z 3 ; *= no CHF N = 27 o CHF N= 3 FIGURE 2. The relationship between pulmonary artery peak systolic pressure and pulmonary to systemic flow ratio in the 57 patients with a residual ventricular septal defect after repair of tetralogy of Fallot. Note the seven patients in the upper left portion of the graph. These patients have pulmonary vascular resistances of greater than 3 mm Hg/L/min/m2. A bbreviation: CHF = congestive heart failure. TABLE 3. Preoperative Pulmonary Artery Pressure and Resistance in Postoperative Patients with Congestive Heart Disease, Pulmonary Artery Hypertension and Pulmonary Vascular Obstructive Disease PAP Pt. No. Preop Postop (mm Hg) PVR/SVR Preop Postop 2-8/16.5/2 3 5/1 5/15 2.3/23.2 / / 9/3 /21 1/ /7 52/27 1/1 3.5/5 18 5/22 71/6 3/16 8./ /9 8/2.5/2 3.1/ /15 55/2 1.6/2 3.2/31 Abbreviations: Pt = patient; PVR = pulmonary vascular resistance; SVR = systemic vascular resistance; preop = preoperatively; postop = postoperatively; PAP = pulmonary artery pressure. due to a very large pulmonary blood flow (i.e., hyperkinetic hypertension) since pulmonary vascular obstructive disease (pulmonary vascular resistance > 3 mm Hg/L/min/m2) was present in 7/19 patients with postoperative pulmonary artery hypertension and a large residual VSD. In at least two of the seven patients, increased pulmonary vascular resistance was already present preoperatively (table 3). Other Residua and Sequelae Left ventricular dysfunction secondary to transsection of an anterior descending coronary artery that arose from a right coronary artery was the cause of CHF in one patient (no. 3). Marked bradycardia (heart rate: 35) due to postsurgical complete heart block resulted in CHF in another patient (no. 32). Following implantation of a pacemaker, the CHF disappeared. Aortic regurgitation was present in /36 patients with congestive heart failure and /66 without CHF. The lesion was hemodynamically significant and contributed to the left ventricular volume overload in only one patient (no. 16), who also had significant subaortic stenosis. Ventricular End-Diastolic Pressure Resting right ventricular end-diastolic pressure was elevated (> 8 mm Hg) in 5/12 patients and left ventricular end-diastolic pressure was elevated (> 1 mm Hg) in 17/12 patients. Clinical congestive heart failure was present in 35 of 3 patients with elevated right ventricular 25 2 RIGHT VENTRICULAR END DIASTOLJC 15 PRESSURE mmhg 1 5 go OO O go Gt.* -;.o * O O( 2 3 N =58 r -.73 o -CHF * No CHF FIGURE 3. The relationship between right ventricular enddiastolic pressure and pulmonary-to-systemic flow ratio in the 58 patients with a residual ventricular septal defect after repair of tetralogy of Fallot. Abbreviation. CHF = congestive heart failure.

5 CHRONIC CHF IN POSTOP TF/Rocchini et al. 39 pressure and in all 17 patients with elevated left ventricular pressure. There is a direct relationship between the magnitude of the residual left to right shunt after repair and the elevation of right and left ventricular end-diastolic pressures (r =.73 and r =.65 respectively; P <.1) (figs. 3 and ). A Qp/Qs > 2 was invariably associated with a left ventricular end-diastolic pressure of > 1 mm Hg and with few exceptions, a right ventricular end-diastolic pressure of > 1 mm Hg. OF N=12 _= CHF Downloaded from by on October 17, 218 Age at Repair Congestive heart failure was present in 16/22 (75%) of patients whose intracardiac repair was performed between 15 and 28 years of age, in 19/69 (28%) patients repaired between 3 and 15 years of age (P <.1) and in 1/11 (9%) patients less than three years of age at repair (fig. 5). However, 13/22 patients greater than 15 years of age had a previous Potts anastomosis, while only 3/69 patients between 3-15 years had a previous Potts shunt. Discussion The inherent bias in the selection of patients for postoperative cardiac catheterization greatly overestimates the incidence of chronic CHF and other residua observed in our patients with TF. However, the hemodynamic observations at cardiac catheterization provided us with an opportunity to gain insight into the etiological factors responsible for the development of chronic CHF and other residua observed in our patients with TF. The residua and sequelae after repair of TF which may lead to chronic CHF are summarized in table. The majority of the patients with chronic CHF have a large residual ventricular septal defect. A left-to-right shunt with a pulmonary-to-systemic flow ratio equal to or greater than 2 was uniformly associated with heart failure (fig. 1). The left ventricle in patients with TF is unable to accommodate the volume overload imposed by the left-to-right shunt.' The left ventricular end-diastolic pressure is elevated and tends to increase with increasing pulmonary flow (fig. ). This may be related, in part, to left ventricular LEFT 2 VENTRICULAR END DIASTOLIC PRESSURE 15 mmhg 5 * so ** * * oo c o 3 N= 58 r.65 o =CHF *=NoCHF FIGURE. The relationship between left ventricular end-diastolic pressure and pulmonary-to-systemic flow ratio in the 58 patients with residual ventricular septal defects after repair of tetralogy of Fallot. Abbreviation: CHF = congestive heart failure. A AGE AT TIME OF REPAIR yrs. FIGURE 5. The effect of age at the time of intracardiac repair of tetralogy of Fallot on the development of congestive heart failure (CHF) in 12 patients. dysfunction resulting from a long standing large systemicto-pulmonary shunt" or chronic cyanotic heart disease."5 These are undoubtedly not the only explanations since some of the patients with a persistent VSD and CHF had a small systemic-to-pulmonary shunt or had a shunt for only a short duration. Furthermore, left ventricular function in cyanotic patients without residual defects, regardless of age at operation, is usually normal.' 16 Some degree of right ventricular myocardial dysfunction produced by akinesis from the ventriculotomy scar or large outflow patch may also lead to the development of CHF when the right ventricle is subjected to additional volume load from a left-to-right shunt. There is a direct relationship between the size of the left-toright shunt and right ventricular end-diastolic pressure (fig. 3). Associated lesions producing volume overload such as tricuspid regurgitation, aortic regurgitation, or the presence of additional pressure load in the right ventricle from unrelieved right ventricle outflow obstruction undoubtedly TABLE. Hemodynamic and Anatomic Abnormalities in 36 Patients with Congestive Heart Failure After Repair of Tetralogy of Fallot No. % Residual ventricular septal defect 3 83 Residual ventricular septal defect and pulmonary stenosis 1 39 Tricuspid regurgitation 1 28 Pulmonary artery hypertension 2 56 Residual aortopulmonary shunt 9 25 Infrequent: Unrecognized muscular ventricular septal defect 2 6 Fracture of calcified Ivalon patch 1 3 Severe residual right ventricular outflow obstruction 2 6 Large aorta to pulmonary artery collateral 1 3 Complete heart block 1 3 Coronary artery injury 1 3 Aortic regurgitation 11 Aortic stenosis 1 3

6 31 CIRCULATION VOL 56, No 2, AUGUST 1977 Downloaded from by on October 17, 218 would increase myocardial work and further aggravate the congestive heart failure. The influence of pulmonary regurgitation on the development of congestive heart failure is unclear. Moderate to severe pulmonary regurgitation was equally distributed among patients with and without CHF (1/36 patients with and 18/66 without CHF). In no patient was pulmonary regurgitation alone responsible for CHF. It is likely, however, that moderate to severe pulmonary regurgitation exacerbates right ventricular overload in patients with tricuspid regurgitation and/or ventricular septal defects. Severe right ventricular outflow tract obstruction in the absence of a large VSD is an uncommon cause of congestive heart failure in the postoperative patients with TF. It occurred in only 2/36 patients. However, significant right ventricular outflow tract obstruction in association with an anatomically large VSD was present in almost 5% of the patients with CHF. Both tricuspid regurgitation and pulmonary artery hypertension are frequently present in postoperative patients with a large residual ventricular septal defect and chronic CHF. Tricuspid regurgitation usually results from intraoperative injury to the tricuspid valve or its attachments, distortion of the valve as a result of anchoring the VSD patch to the septal leaflet and less frequently as a result of myocardial dilatation or marked volume overload. As previously noted by others'"13 the increase in pulmonary artery pressure was related to the Qp/Qs and the presence of a previous longstanding systemic to pulmonary artery shunt. Pulmonary vascular obstructive disease contributed to the elevation of pulmonary artery pressure in seven of our patients (table 3). The frequency of CHF after repair of TF was shown to increase with age. This is, at least in part, related to adverse effects from prior palliative procedures in the majority of older children and young adults. The systemic to pulmonary shunts may remain open after repair,"7 result in the development of pulmonary vascular obstructive disease or left ventricular dysfunction.9' 11, 17 The increased incidence of congestive heart failure and the other complications of palliative surgery'8 are some of the major reasons for recommending primary and early repair whenever technically feasible.'9 2 Review of the catheterization findings revealed some unusual causes for congestive heart failure in patients with repaired TF. Two patients had an additional muscular VSD and in one patient, congestive failure developed after fracture of a 5-year-old Ivalon patch. Complete heart block, left ventricular dysfunction secondary to coronary artery transsection, a large residual systemic to pulmonary artery collateral and aortic stenosis, each caused CHF in one patient. Our findings indicate that the presence of chronic congestive heart failure in postoperative patients with TF requires a thorough bilateral heart catheterization and angiography. A surgical correctable lesion, usually a residual VSD, is nearly always present but must be clearly defined preoperatively. The presence in some patients of an additional muscular ventricular septal defect, patent systemic to pulmonary artery shunt, aortic or tricuspid regurgitation emphasizes the importance of accurately identifying the location of the left to right shunt and other hemodynamic lesions prior to another surgical intervention. Fortunately, reoperation in patients with TF does not carry an excessive risk and its results are excellent.2' Delay in recommending reoperation may lead to further cardiac decompensation, increasing symptoms, the development or progression of pulmonary vascular obstructive disease and increasing myocardial dysfunction. References 1. Kirklin JW, Wallace RB, McGoon DC, DuShane JW: Early and late results after intracardiac repair of tetralogy of Fallot. Ann Surg 162: 578, Kirklin JW, Karp RB: The tetralogy of Fallot from a surgical viewpoint. Philadelphia, WB Saunders, Burnell RH, Woodson RD, Lees MH, Bristow JD, Starr A: Results of correction tetralogy of Fallot in children under four years of age. J Thorac Cardiovasc Surg 57: 153, Goldman BS, Mustard WT, Trusler GS: Total correction of tetralogy of Fallot: Review of ten years experience. Br Heart J 3: 563, Gotsman MS, Beck W, Barnard CN, O'Donovan TG, Schrire V: Results of repair of tetralogy of Fallot. Circulation : 83, Ruzyllo W, Nihill MR, Mullins CE, McNamara DG: Hemodynamic evaluation of 221 patients after intracardiac repair of tetralogy of Fallot. Am J Cardiol 3: 565, Lillehei CW, Levy MJ, Adams P, Anderson RC: Corrective surgery for tetralogy of Fallot: Longterm follow-up by postoperative recatheterization. J Thorac Cardiovasc Surg 8: 556, LaFarge CG, Miettinen OS: The estimation of oxygen consumption. Cardiovasc Res : 23, VonBermuth G, Ritter DG, Frye RL, Weidman WH, Davis GD, McGoon DC: Evaluation of patients with tetralogy of Fallot and Potts anastomosis. Am J Cardiol 27: 259, Paul MH, Miller RA, Potts WJ: Longterm results of aortic-pulmonary anastomosis for tetralogy of Fallot: An analysis of the first 1 cases 11 to 13 years after operation. Circulation 23: 525, Roisman ML, Beller BM, O'Keefe JD: Irreversible pulmonary hypertension after correction of tetralogy of Fallot. Chest 62: 3, Kinsley RH, McGoon DC, Danielson GF, Wallace RB, Moser DD: Pulmonary artery hypertension after repair of TF. J Cardiothorac Surg 67: 11, Rosenthal A, Sonnenblick E, Lesch M: Postoperative Congenital Heart Disease. New York, Grune and Stratton, Inc, Rocchini A, Keane JF, Freed MD, Castaneda AR, Nadas AS: Left ventricular function after surgical repair of tetralogy of Fallot. Am J Cardiol 37: 167, Jarmakani J, Graham J, Canent R, Jewett P: Left heart function in children with tetralogy of Fallot before and after palliative or corrective surgery. Circulation 6: 78, LaCorte M, Williams R: The echocardiographic assessment of left ventricular function following tetralogy of Fallot repair. Am J Cardiol 35: 15, Morriss J, McNamara D: Residual sequelae and complications ofsurery for congenital heart disease. Prog Cardiovasc Dis 28: 35, Greenwood RD, Rosenthal A, Nadas AS, Freed MD, Castaneda AR: Ascending aorta-pulmonary artery anastomosis for cyanotic congenital heart disease. Am Heart J (in press) 19. Rosenthal A: When to operate on congenital heart disease. In Controversy in Cardiology, edited by Chung EK. New York, Springer-Verlag, Castaneda AR, Williams RG, Rosenthal A, Sade RM: Tetralogy of Fallot: Primary repair in infancy. In Symposium in Congenital Heart Disease, edited by Keith, Rowe. Toronto, Hospital for Sick Children (in press) 21. Castaneda AR, Sade R, Lamberti J, Nocoloff D: Reoperation for residual defects after tetralogy of Fallot repair. Surg 76: 11, 197