Hemodynamics and intracardiac conduction after operative repair of tetralogy of fallot

Hemodynamics and Intracardiac Conduction After Operative Repair of Tetralogy of Fallot DOLORES TAMER, MD, GRACE S. WOLFF, MD, PEDRO FERRER, MD, ARTHUR S. PICKOFF, MD, ALFONSO CASTA, MD, ASHOK V. MEHTA, MD, OTTO GARCIA, MD, and HENRY GELBAND, MD

Electrophysiologic studies were performed in 47 children aged 3 to 18 years, 15 of whom had cardiac arrhythmias 1 to 15 years after repair of tetralogy of Fallot. Six exhibited sinus or atrioventricular nodal dysfunction, 8 had ventricular extrasystoles, and 1 had supraventricular tachycardia. Hemodynamic and electrophysiologic data were obtained at postoperative catheterization. Although electrophysiologic responses were abnormal in a proportion of both the children with and those without arrhythmia, hemodynamic values were similar. Three of 6 children with impaired sinus impulse generation

or atrioventricular nodal conduction had a prolonged A-H interval, and in 3 Wenckebach heart block developed at low pacing rates. Ventricular ectopic rhythm was not associated with any particular abnormality of basic intracardiac conduction intervals. Thus, arrhythmias and conduction abnormalities are not consistently related to residual right ventricular hypertension. Abnormalities in electrophysiologic function are common after repair of tetralogy of Fallot in patients with sinus rhythm and may have prognostic implications for these patients.

Tetralogy of Fallot is a common congenital heart defect for which reparative open heart surgery has been performed in thousands of patients over the past 2 decades. The immediate risk of the surgery has steadily declined; however, a small but alarming incidence of late sudden u n e x p e c t e d deaths represents a continuing source of concern. Bifascicular block, TM transient complete heart block, 4,5 ventricular ectopic r h y t h m , 6 especially with right ventricular hypertension, 7-9 and late development of left axis deviation 1° have been associated with sudden u n e x p e c t e d death. This investigation a t t e m p t s to d e t e r m i n e whether intracardiac hemodynamics relate to the intracardiac conduction properties in children after repair of tetralogy of Fallot. Also, because a r r h y t h m i a s have been implicated in late mortality, it was hoped to determine

whether those children with arrhythmias have specific conduction abnormalities t h a t m a y serve to identify those requiring increased surveillance. T h e study was begun in 1975 to collect electrocardiographic information and to correlate this with postoperative hemodynamics and electrophysiologic functions at cardiac catheterization.

Methods Forty-seven children were studied after operative repair of tetralogy of Fallot. None had arrhythmia preoperatively on standard electrocardiography. In 13 children the indications for admission included findings of residual heart disease or electrocardiographic abnormality; the remaining 34 were admitted electively. There were 24 male and 22 female patients, aged 3 to 18 years. Surgery had been performed 1 to 15 years previously (average 4.3). All children had a routine 12-lead electrocardiogram. After informed consent had been obtained, cardiac catheterization was performed and included hemodynamics, cineangiography, and determination of basic intracardiac conduction intervals, right ventricular activation time, and sinus node recovery time to rapid atrial pacing. Thirty-one patients had programmed atrial extrastimulation studies, and 21 had programmed ventricular stimulation. Thirty patients also had

From the University of Miami School of Medicine, Department of Pediatrics, Miami, Florida. This study was supported in part by Grant HL07436 from the National Institutes of Health, Bethesda, Maryland. Manuscript received December 9, 1981; revised manuscript received August 19, 1982, accepted August 20, 1982. Address for reprints: Dolores Tamer, MD, Pediatric Cardiology, Department of Pediatrics, University of Miami School of Medicine, P.O. Box 016820, Miami, Florida 33101. 552

February 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51

TABLE I

553

Arrhythmias in 15 Patients After Operative Repair of Tetralogy of Fallot (Group A)

Case

Age (yr) at Op

Age (yr) at Study

ECG Axis

Rhythm

RV Pressure (mm Hg)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

5 3 3 4 3 2 4 7 6 7 5 6 5 4 3

10 7 6 6 4 4 5 18 12 17 7 11 12 8 7

0 ° RBBB -t-120°RBBB - 2 0 ° RBBB - 3 0 ° RBBB ,1,1,120° RBBB Indet RBBB + 1 6 0 ° RBBB - 3 0 ° RBBB 0 ° RBBB - 3 0 ° RBBB + 1 4 0 ° RBBB Indet RBBB ,1,75° RBBB + 9 0 ° RBBB ,1,110 ° RBBB

Sinus arrest, Mobitz II Sinus arrest, junctional bradycardia Spontaneous Wenckebach Spontaneous Wenckebach Junctional bradycardia, P-R = 0.22 s, PVC Sinus and junctional bradycardia PVC, bigeminy PVC, bigeminy PVC, bigeminy PVC PVC, PAC PVC, PAC PVC, PAC PVC, PAC SVT

44/4 32/6 102/7 56/8 90/8 24/6 66/3 40/6 20/4 44/10 80/10 36/12 46/4 28/4 34/6

* PVC = ventricular extrasystoles; RBBB = right bundle branch block; RV = right ventricular; SVT = supraventricular tachycardia.

ambulatory electrocardiographic monitoring and 29 had exercise stress testing using the modified Bruce protocol. The materials and methods used for catheterization and electrode placement have previously been described, n Agents used for sedation included Demerol®, Phenergan ®, Thorazine®, and pentobarbital sodium. Basic intervals and apical activation were compared with those of age-matched preoperative children studied in this laboratory. 12 Sinus node recovery time was measured from the last paced atrial deflection to the first spontaneous atrial beat and then divided by the resting sinus cycle length and expressed as a percentage. The majority of preoperative children have values for this percentage <160% in our laboratoryJ 2 The response to rapid atrial pacing was monitored for development of Wenckebach heart block. The cycle length at which Wenckebach periodicity occurred was compared with the reference distribution for our laboratoryJ 2 Values for Wenckebach periodicity were considered abnormal if they fell outside this reference. Definitions of basic intervals, right atrial effective refractory period, atrioventricular nodal effective refractory period, right ventricular effective refractory period, and right ventricular apical activation time have been cited 13 previously.

TABLE II

E l e c t r o c a r d i o g r a p h i c f i n d i n g s : Various a r r h y t h mias were discovered by scalar electrocardiography (11 cases), a m b u l a t o r y monitoring (3 cases), or exercise stress testing (1 case), The arrhythmias are listed in T a b l e I and characterize Group A. Six children had various manifestations of sinus or atrioventricular nodal dysfunction, including sinus arrest, spontaneous Mobitz type I or II atrioventricular block, and junctional bradycardia (Table I). One child had supraventricular tachycardia and 8 had ventricular ectopic r h y t h m , which consisted of uniform single ectopic beats or bigeminy that were suppressed or unaffected by exercise. T h e r e were 32 children without a r r h y t h m i a and these comprise Group B (Table III). T h e average age of children in Group A was 8.9 years, of Group B 7.7 years, and of the s t u d y group as a whole 8.1 years. Twelve-lead scalar electrocardiograms revealed right bundle branch block in 96% of all cases and in all 15 cases in Group A.

Electrophysiologic Data: Group A SCL

A-H (ms)

H-V (ms)

1 2 3

8t8 500 600

126" 66 88

50 57"(54) 26

165 220 215

4 5 6 7 8 9 10 11 12 13 14

680 625 645 590 740 816 850 610 748 770 522

230" 15t* 102 80 65 70 76 80 90 105 69

41 39 35 42 48 35 65 * (55) 58*(54) 49 44 44

ND 160 210 220 ND 230 ND ND 280" 190 220

15

706

85

30

240"

Case

Results

RAERP (ms)

AVN ERP

RV ERP

SNRT (%)

S1S1 at Wenckebach

PRBBB

410" AL F = 570 S = 450 ND AL 400 275 ND 390 ND ND AL 350 F = 295 S=AL 300

245 ND 270

118% 170 % * 105%

None 300 470*

-I-+

ND 160 ND ND ND ND ND ND 240 ND 220

140 % 208" % 138 % 120% 148% 166" % 135 % 108% 150 % 157 % 153%

500" 330 500" 310 0 410" 0 350 350 350 0

+ --+ ---+ ---

220

129 %

350

+

* Abnormal valve; parentheses indicate mean value + 2 standard deviations. AL = atrial limited; apex = RV apex pacing; AVN = atrioventricular nodal; DAVN = dual atrioventricular nodal pathways; ERP = effective refractory Period; F = fast atrioventricular nodal pathway; LAD = left axis deviation; ND = not done; Op = operation; out = RV outflow pacing; P -- paced; PAC = atrial extrasystoles; PRBBB = proximal right bundle branch block; preop = preoperative; RA = right atrial; S = slow atrioventricular nodal Pathway; SCL = sinus cycle length; SNRT = sinus node recovery time; $1S1 = interstimulus interval; other abbreviations as in Table I.

554

HEMODYNAMICSAND CONDUCTION

TABLE III

Hemodynamic and Electrophysiologic Data in Arrhythmia-Free Patients (Group B)

Case

Age (yr) at Op

Age (yr) at Study

RVP (mm Hg)

16 17

5 3

13 4

48/8 30/8

18

3

7

26/7

19 20 21 22

3 5 1 5

18 10 5 9

27/6 44/14 58/2 20/6

23

2

3

48/6

24

6

17

42/8

Intervals (ms) SCL (ms)

P-A

A-H

880 500P 615 900

30

60

16 15

63 102

608 750 720 814 600P 610

16 20 19 29

72 60 40 89

42 40 47 39

22

45

25 18

25

2

7

34/6

500P 810 540

26

5

11

24/7

720

27

3

4

17/5

390P 478 450P 557 580 480P

H-V 30

ERP (ms) RA

260

118

320

290* ND

117 150

None 350

205 3O5 AL AL 320 30O

260 ND 410" 370* 290 ND

128 139 160" 150

None 345 365* 480*

35

250* 190 230 270 250* 260*

160"

None

83 93

50 40

260* 175 220

AL AL AL

135 171"

300 None

30

62

35

200

AL

220 290 270 out 230 apex 250

138

200

PRBBB, split His

25

65

40

None

DAVN

60 80

44 30

190

260 = F 240 = S 3OO 3OO AL AL F = 430 S = 335 S = 315 S = 300 F = 310 S = AL F = 290 S = AL AL

104 113

39 25

145 140 250 165 220 190

240

AL

41 50*(45)

4

39/4

29

3

5

33/8

30

4

6

40/8

680

29

70

50*(45)

180

31

3

4

32/6

693

21

69

45

200

490P 4

200

250

121

400"

310

136

290

156

330

225

23 20 18

70 90 50

40 45 40

150 100 110

375* None None

800 490 620 540 720 840 740 638 680 920 674 714

21 20 25 10 18 21 13 24 12 18 29 22

77 60 84 60 70 80 62 42 70 68 80 90

45 64*(49) 55*(49) 25 60*(57) 35 35 37 55 42 30 35

138 156 120 153 164" 133 100 103 100 83 125 120

375* 300 None 333 None None None None None None 375* 428*

38/8 84/12 68/16

36 37 38 39 40 41 42 43 44 45 46 47

3 3 2 3 3 6 5 5 3 6 5 5

7 5 4 5 7 10 13 10 8 13 9 8

42/5 40/8 56/12 36/12 50/10 36/10 30/8 108/10 38/10 36/4 24/2 60/13

36

480* None

400P 620 680 560

8 5 4

78

139

541

4 3 1

0

220 260 230 out 190 apex 320* ..

20/4

33 34 35

Other

AL AL 290 AL

2

1

RV

S1Slat Wenckebach (ms)

260* 180 160 290*

28

32

AVN

SNRT as % SCL

300 out 280 apex

LAD,"I)RBBB, split His PRBBB, split His PRBBB PRBBB LAD preop, PRBBB AVN, reentry Split His

PRB'I313, LAD Multiple AVN pathways DAVN, AVN reentry Ventricular reentry PRBBB LAI3," ' PRBBB

PRB'I31~ PRBBB PRBBB + LAD

LAd preop

PRggB

LAD = left axis deviation; ND = not done; PRBBB = proximal right bundle branch block; other abbreviations as in Tables I and II

Four children in Group A and 4 in Group B had left axis deviation that was not present preoperatively. Hemodynamic findings: The majority of children had good hemodynamics postoperatively (Tables I and III), although mild pulmonary valve stenosis or pulmonary insufficiency, or both, were common. Four children in Group A and 4 in Group B had severe pulmonary stenosis, with right ventricular pressure >60 mm Hg. Two children in Group A and 1 in Group B required reoperation after these studies. Tricuspid valve regurgitation was found on right ventricular cineangiography in 4 children each in Groups A and B. Right atrial mean pressure was increased >8 mm Hg in 1 child in Group A (Case 11) and 2 in Group B. Right ventricular end-diastolic pressure was increased (>8 mm Hg)

in 3 children in Group A and 10 children in Group B, and may reflect residual pulmonary insufficiency, right ventricular dysfunction, or decreased right ventricular compliance. Right ventricular aneurysm was defined angiographically by dilatation of the right ventricular outflow area to twice the diameter of the main pulmonary artery as visualized on the frontal view and was present in 2 children in Group A and 9 in Group B. Pulmonary stenosis was considered mild to moderate when right ventricular pressure was <60 mm Hg. Significant pulmonary insufficiency was defined by at least a grade 2/6 diastolic murmur at the pulmonary area. Three children had small residual ventricular septal defects. The residual cardiac defects are summarized in Table IV.

February 1983

TABLE IV

Residual Cardiac Lesions After Operative Repair of Tetralogy of Fallot (47 Cases)

THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51

TABLE V

Abnormal Intracardiac Conduction Findings Group A

Group A

Significant PS (RVSP > 6 0 mm Hg) Significant PI Significant TI Mild/moderate PS Residual VSD RV aneurysm Total

555

Group B

Group B

n

%

n

%

4 9 4 8 1 2 15

27 60 27 53 0.07 13 ...

4 26 4 16 2 9 32

13 81 13 50 13 28 ...

No significant differences on chi-square analysis. PI = pulmonary insufficiency; PS = pulmonary stenosis; RV -- right ventricular; RVSP = right ventricular systolic pressure; TI = tricuspid insufficiency; VSD = ventricular septal defect.

Electrophysiologic data: These data are similar to conduction studies reported previously 14 16 after operative repair of tetralogy of Fallot. In addition, similar percentages of abnormal electrophysiologic findings were present in Group A as a whole and Group B (Tables II, III, and V). However, when the subgroup of cases of arrhythmia with impaired sinus or atrioventricular nodal conduction was analyzed separately, a higher aggregate number of cases had abnormal conduction values as compared with cases with ventricular ectopic rhythm or patients in Group B. There was no single pattern or abnormal measurement of conduction in the 6 children with impaired function (cases 1 to 6), although each had at least 1 abnormal value in the studies performed (Table II). Three children had a prolonged A-H interval of atrioventricular nodal dysfunction (cases 1, 4, and 5). One had a prolonged H-V interval consistent with His-Purkinje disease (case 2). In 3 children Wenckebach heart block developed at abnormally low paced rates (cases 3, 4, and 6) and in case 3 the block occurred below the His bundle. Only 2 of the 4 children (cases 2 and 5) suspected of having sinus node dysfunction had prolongation of the sinus node recovery time, suggesting either that this measurement is not sensitive, that vagal slowing may account for the observed bradycardia, or that the circumstances of the study did not reflect the clinical situation because of sedation or other factors. Discussion

Fifteen of 47 patients with postoperative tetralogy of Fallot manifested significant cardiac arrhythmias. In the 6 children with impaired sinus generation or atrioventricular nodal conduction, various electrophysiologic abnormalities were noted including prolonged A-H or H-V interval, increased sinus node recovery time, atrioventricular nodal effective refractory period, or appearance of Wenckebach periodicity at low pacing rates. Atrial and ventricular effective refractory periods Were normal in these children. More children in Group A had a prolonged A-H interval (p = 0.05); however, the frequency of the other abnormalities was not statistically different in Groups A and B. Specifically, H-V interval, sinus node recovery time, Wenckebach periodicity, and atrial, ventricular, and atrioventricular

SNRT > 1 6 0 % Wenckebach at S1S~ > 3 6 5 ms A-H over 2 SD for age H-V over 2 SD for age Proximal RBBB LAD -I- PRBBB RAERP over 1 SD for rate RVERP over 1 SD for rate

n

%

n

%

3/15 4/15 3/15 3/15 6/15 4/15 2/11 0/6

30 26 20* 20 40 27 18 0

4/32 8/32 0/32 5/32 13/32 4/32 6/20 4/15

13 25 0 16 32 13 30 27

* p = 0.05 (chi-square).

nodal effective refractory periods were found in similar proportions in patients with and without overt arrhythmias. Children with ventricular ectopic rhythm had no consistent abnormality of basic conduction intervals. In 5 of 6 cases in Group B in which comparison was possible, the right ventricular effective refractory period was longer at the outflow region of the right ventricle than at the apex. Only 2 of the 8 children cases with extrasystoles had ventricular stimulation. Provocative measures such as multiple stimuli or burst ventricular pacing might have uncovered more abnormalities. Ventricular tachycardia was reported after repair of tetralogy of Fallot 17 which used a reentry circuit around the margin of the ventriculotomy. Kavey et al is noted that fewer patients in their series of cases of ventricular tachycardia had outflow patches. One may speculate that the interposition of fabric may tend to interrupt a potential ventricular reentrant pathway. The difference we observed in ventricular refractoriness may be potentially arrhythmogenic. Other potentially arrhythmogenic factors such as prolonged effective refractory period, presence of dual atrioventricular nodal pathways, and split His potentials (reported previously) 11,19 are of uncertain prognostic significance. In addition, the large proportion of children in Group B who had sinus rhythm and abnormal intracardiac conduction values may have latent conduction disease and will require continued medical follow-up to determine whether impaired conduction or ventricular arrhythmia appears. 2° Despite the findings of other investigators that ventricular extrasystoles were more frequent when right ventricular hypertension persisted, 7-9 our data show no consistent relation. Two patients with impaired conduction and 2 with ventricular extrasystoles had a right ventricular systolic pressure >60 mm Hg, but the majority of patients with arrhythmia had good hemodynamic results. Although a direct relation between the rate of pacemaker depolarization and myocardial stretch was demonstrated in vitro, 21the present findings do not support a mechanical factor as the likely mechanism in clinical cases of postoperative rhythm disturbances. Furthermore, the reported series of ventricular tachycardia17,1s,22 occurred in cases with no significant residual pulmonary stenosis. The increase in ventricular arrhythmias noted by James and co-workers 22 after exercise may represent increased ventricular excitability

556

HEMODYNAMICS AND CONDUCTION

due to sympathetic discharge in patients who have a latent reentry circuit around the ventriculotomy. The observation that several patients with ventricular extrasystoles and right ventricular hypertension had sudden death 7-9 may reflect the poor tolerance of patients with impaired hemodynamic function to the circulatory effects of arrhythmias. Ventricular ectopic rhythm or conduction impairment, residual hemodynamic stress, and myocardial injury may combine to bring about sudden death. Furthermore, the incidence of ventricular arrhythmia was increased in older patients especially in those with tetralogy of Fallot followed up 10 years or more, 2,6,9,1°,22-24 suggesting a relation to late fibrosis in the myocardiumo Recent methods of myocardial preservation during surgery and the practice of operating at an earlier age may change the future incidence of ventricular arrhythmias. Conclusions from these data are threefold: (1) children with evidence of Mobitz type I or II atrioventricular block usually have electrophysiologic abnormalities; (2) right ventricular hypertension or residual heart disease does not correlate well with the occurrence of arrhythmias or with electrophysiologic abnormalities; (3) a large number of children after repair of tetralogy of Fallot have conduction abnormalities and normal sinus rhythm. Although the arrhythmias and the electrophysiologic findings were well tolerated in this series and in others, 15,23 a longer period of observation is necessary before the prognostic significance of these findings is known. Acknowledgment: We thank Diane Mullen for her secretarial assistance and Ralph Portuondo for technical assistance.

3. 4. 5. 6. 7. 8. 9. 10. 11.

12.

13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

References 1. Wolff GS, Rowland TW, Elliso n RC. Surgically induced right bundle branch block with left anterior hemiblock. An ominous sign in postoperative tetralogy of Failer. Circulation 1972;46:587-594. 2. Quaftlebaum TG, Varghese DJ, Neill CA, Donahoo JS. Sudden death among

23. 24.

postoperative patients with tetralogy of Fallot. A follow-up study of 243 patients for an average of twelve years, Circulation 1976;54:289-293. Downing JW, Kaplan S, Bone KE. Post-surgical left anterior hemiblock and right bundle branch block. Br Heart J 1972;34:263-270. GodmanMJ, Roberts NK, Izukawa T. Late postoperative conduction disturbances after repair of ventricular septal defect and tetralogy of Fallot. Analysis by His bundle recordings. Circulation 1974;49:214-221. Krongrad E. Prognosis for patients with congenital heart disease and postoperative intraventricular conduction defects. Circulation 1978;57: 867-869. James FW, Kaplan S, Chou TC. Unexpected cardiac arrest in patients after surgical correction of tetralogy of Fallot. Circulation 1975;52:691-694. Garson A, Gillette PC, Gutgesell HP, McNamara DG. Stress-induced ventricular arrhythmla after repair of tetralogy of Fallot. Am J Cardiol 1980;46:1006-1012. Gillette PC, Yeoman MA, Mullins CE, McNamara DG. Sudden death after repair of tetralogy of Fallot. Circulation 1977;56:566-571. Fuster V, McGoon DC, Rifler DG. Factors affecting the long term results (11-21 years) of open heart surgery for tetralogy of Fallot (abstr). Am J Cardioi 1979;43:362. Deanfield JE, McKenna WJ, Hallidie-Smlth KA. Mechanism of sudden death in postoperative tetralogy of Fallot (abstr). Br Heart J 1979;42:235-236 Casta A, Wolff GS, Mehta A, Tamer D, Garcia OL, Pickoff AS, Ferrer PL, Sung RJ, Gelband H. Dual atrioventricular nodal pathways: a benign finding in arrhythmia-free children with heart disease. Am J Cardiol 1980;46: 1013-1018. Wolff GS, Casta A, Kaiser G, Pickoff AS, Mehta AV, Tamer D, Garcia OL, Ferrer PL, Smith K, Gelband H. Sinus and atrioventricular nodal function: preoperative and early postoperative assessment in children. J Thorac Cardiovasc Surg 1982;83:141-145 Roberts N, Gelband H. Cardiac Arrhythmias in the Neonate, infant and Child. New York: Appleton-Century-Crofts, 1977:151. Neches WH, Parle SC, Mathews RA, Lenox CC, Marin-Garcla J, Zuberbuhler JR. Tetralogy of Fallet: postoperative electrophysiologic studies. Circulation 1977;56:713-719. Pahlajani DB, Serratto M, Mehta A, Miller RA, Hastreiter A, Rosen KM. Surgical bifascicular block. Circulation 1975;52:82-87. Levin AR, Haft JI, Engle MA, Ehlers KH, Klein AA. Intracardiac conduction intervals in children with congenital heart disease. Circulation 1977;55. 286-294. Horowitz LN, Vetter VL, Harken AH, Josephson ME. Electrophysiologic characteristics of sustained ventricular tachycardia occurring after repair of tetralogy of Fagot. Am J Cardiol 1980;46:446-452. Kavey REW, Blackman MS, Sondheimer HM. Ventricular dysrhythmias after repair of tetralogy of Fallot (abstr). Circulation 1980;62:111:1!1-72. Wolff GS, Tamer D, Garcia OL, Ferrer PL, Pickoff AS, Sung RJ, Gelband H. His-Purkinje conduction findings after cardiac surgery in children. Circulation 1980;62:615-621. Pickoff AS, Mehta AV, Casta A, Ferrer PL, Wolff GS, Tamer DL, Garcia OL, Gelband H. Echocardiographicassessment of right bundle branch injury following repair of tetralogy of Fallot. Circulation 1981 ;63:174-181. Blinks JR. Positive chronoiropic effect of increasing right atrial pressure in the isolated mammalian heart. Am J Physiol 1956;186:299-303. James FW, Kaplan S, Schwartz DC, Chou TC, Sandier ET, Naylor V. Response to exercise in patients after total surgical correction of tetralogy of Fagot. Circulation 1976;54:671-679. Wessel HA, Paul MH, Berry TE, Cole RB, Muster AJ. Prognostic significance of arrhythmia in tetralogy of Fallot after intracardiac repair. Am J Cardiol 1980;46:843-848. Rocchlnl AP. Hemodynamic abnormalities in response to supine exercise in patients after operative correction of tetralogy of Fallot after early childhood. Am J Cardiol 1981;48:325-330.