Rest and exercise ventricular function in adults with congenital ventricular septal defects

Rest and Exercise Ventricular Function in Adults With Congenital Ventricular Septal Defects GEORGE JABLONSKY, MD, J. DAVID HILTON, MD, PETER P. LIU, MD, JOHN E. MORCH, MD, MAURICE N. DRUCK, MD, BEN-ZION BAR-SHLOMO, MD, and PETER R. McLAUGHLIN, MD, With the technical assistance of KATHY M. WINTER, BSc (N)

Rest and exercise right and left ventricular function were compared using equilibrium gated radionuclide angiography in 19 normal sedentary control subjects (mean age 28 years, range 22 to 34) and 34 patients with hemodynamically documented congenital ventricular septal defect (VSD) (mean age 27 years, range 20 to 40). The 34 patients with VSD were divided into 3 groups: those in Group 1 (17 patients) had pulmonary to systemic blood flow ratios of less than 2 to 1; those in Group 2 (12 patients) had prior surgical closure of VSD (mean interval from surgery 17 years, range 9 to 22), and those in Group 3 (5 patients) had Eisenmenger's complex. Gated radionuclide angiography was performed at rest and during each level of graded supine bicycle exercise to fatigue. Heart rate, blood pressure, maximal work load achieved, and right and left ventricular ejection fractions were assessed. The control subjects demonstrated an increase in both the left and right ventricular ejection fractions with exercise (0.70 ±

0.07 to 0.79 4- 0.05 and 0.46 4- 0.06 to 0.57 ± 0.04; p <0.001 for left and right ventricles, respectively). All study groups failed to demonstrate an increase in ejection fraction in either ventricle with exercise. Furthermore, resting left ventricular ejection fraction in Groups 2 and 3 was lower than that in the control subjects (0.59 ± 0.09 and 0.54 4- 0.06 versus 0.70 -t- 0.07; p <0.001) and resting right ventricular ejection fraction was lower in Group 3 versus control subjects (0.30 4- 0.07 versus 0.46 ± 0.06; p <0.001). Thus (1) left and right ventricular function on exercise were abnormal in patients with residual VSD as compared with control subjects; (2) rest and exercise left ventricular ejection fractions remained abnormal despite surgical closure of VSD in the remote past; (3) resting left and right ventricular function was abnormal in patients with Eisenmenger's complex; (4) lifelong volume overload may be detrimental to myocardial function.

Few data are available on the effects of long-term volume overload on right and left ventricular function. 1 However, patients with patent congenital VSD reaching adulthood provide an ideal opportunity to assess this effect because of chronic biventricular volume overload. Rest and exercise gated equilibrium radionuclide angiography has become a widely accepted noninvasive technique for the evaluation of ventricular performance. Wackers et al 2 and Maddahi et al 3 validated the reli-

ability and reproducibility of this technique for the left and right ventricles, respectively. This study was undertaken to evaluate ventricular performance in adult patients with isolated VSD using gated radionuclide angiography. The result is then compared with (1) age-matched normal control subjects, (2) patients with VSD who had previously undergone surgical closure, and (3) patients with Eisenmenger's complex.

From the Division of Cardiology, Department of Medicine, Toronto General Hospital, Toronto, Ontario, Canada. This work was supported by Grant T1-7 from the Ontario and Canadian Heart Foundations, Toronto, Ontario, Canada. Manuscript received February 22, 1982; revised manuscript received August 4, 1982, accepted August 6, 1982. Address for reprints: Peter R. McLaughlin, MD, Cardiovascular Unit, Toronto General Hospital, 200 Elizabeth Street, Toronto, Ontario, Canada, M5G 1L7.

Methods Patient population: Thirty-four patients with previous angiographically proven isolated VSD were selected for study from our adult congenital heart clinic follow-up population. All patients were aged 20 to 40 years with no known history of coronary artery disease or other concurrent forms of heart disease. The patients were divided into 3 groups. 293

294

VENTRICULARFUNCTION IN VENTRICULAR SEPTAL DEFECT

TABLE I

Clinical Characteristics and Gated Nuclear Angiographic Results in Patients With Patent Small Shunts (Group I) Ejection Fraction

Patient

Age (yr)

Shunt Size at Last Cath

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

20 20 20 21 21 21 22 24 25 26 28 28 29 29 29 31 39

1.3:1 1.4:1 1.8:1 1.3:1 1.2:1 1.5:1 1.5:1 1.6:1 1.1:1 1.2:1 1.8:1 1.4:1 1.3:1 1.2:1 1,9:1 1.1:1 1.8:1

Functional Class

Level of Physical Conditioning

LV (Rest/Ex)

RV (Rest/Ex)

Peak Work Load (kpm/kg)

Peak HR X SBP (X 103 mm Hg/min)

Active Active Active Active Active Sedentary Active Active Active Sedentary Active Sedentary Sedentary Active Active Active Active

81/79 64/60 68/69 66/63 70/68 76/70 60/60 58/58 63/66 76/69 70/65 71/72 65/72 64/62 81/79 81/66 76/63

47/50 35/42 40/41 46/46 51/55 44/44 40/53 53/45 32/48 43/47 28/30 46/42 48/53 40/41 55/57 56/42 52/41

12.6 9,2 15.7 10.3 8.1 8.1 10.8 11.9 13.3 8.8 9.0 7.5 3.7 11.0 6.3 9.5 9.3

36.0 19.5 27.0 22.4 21.6 18.0 20.0 34.2 25.5 17.3 15.0 18.0 22.4 22.8 18.4 39.1 25.0

Cath = catherization; Ex = exercise; LV = left ventricle; RV = right ventricle; SBP = systolic blood pressure.

Group 1: This group included 17 patients (9 male and 8 female) who had persistent small left to right shunts without the necessity of surgical correction (Table I). Their mean age was 25 years (range 20 to 39). All had undergone at least 1 cardiac catheterization. The mean time lapse from the last catheterization to the present was 12 years (range 0 to 18). The mean pulmonary to systemic blood flow ratio at the time of the last catheterization was 1.4 to 1. Mean pulmonary artery pressure was normal at 16.7 :t: 3.5 mm Hg (mean 4- standard deviation [SD]). All patients were in New York Heart Assoclarion (NYHA) functional class I. Thirteen of 17 patients were physically active in sports on a regular basis, whereas 4 were considered sedentary. None of the patients was participating in any physical fitness training program. On physical examination all patients had findings typical of VSD. Group 2: This included 12 patients (5 male and 7 female) who had undergone surgical closure of VSD 9 to 22 years (mean 17 years) before the study (Table II). Their mean age was 26 years (range 20 to 33 years). The operation was performed before the advent of present-day techniques of cardioplegia and myocardial protection. All patients were also in NYHA class I. However, only 5 of 12 patients were active

TABLE II

physically. The remainder led voluntarily sedentary lives without participation in regular sports activities. Group 3: This included 5 patients (1 male and 4 female) with Eisenmenger's complex documented at previous cardiac catheterization. Their mean age was 33 years (range 24 to 40). All were variably symptomatic with dyspnea on exertion. Two patients were in NYHA class II, the remainder in class III. Control: The control group consisted of 19 healthy volunteers (15 male and 4 female) with a mean age of 28 years (range 22 to 34). This was comparable to the mean age of 27 years (range 20 to 40) in our overall study group. All of the volunteers had a normal cardiac history and normal results on physical examination. Protocol: All patients had a complete history, physical examination, standard 12-lead electrocardiogram, posteroanterior and lateral chest X-ray, and supine multiple-gated equilibrium cardiac blood pool scintigrams both at rest and with graded bicycle exercise to fatigue. R a d i o n u c l i d e angiograms: Equilibrium gated radionuclide angiograms were obtained by imaging the cardiac blood pool in the 45 ° left anterior oblique position with a 10 ° cranial

Clinical Characteristics and Gated Nuclear Angiographic Results in Patients With Postoperative Ventricular Septal Defect (Group II)

Patient

Age (yr)

Date of Last Op

Age (yr) at Last Op

Cardiac Operations (n)

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

20 20 21 24 25 26 26 27 29 29 32 33

1963 1961 1961 1959 1960 1961 1960 1965 1965 1965 1959 1970

3 1 2 4 5 9 6 13 6 15 12 24

1 1 1 1 1 1 1 1 1 2 3 1

Shunt Size Before Operation 3.2:1 ... 2".5:'1 ... 3'.0:'1 3.0:1 1".4:'1 2.3:1 2.5:1

Ex = exercise; LV = left ventricle; Op = operation; RV = right ventricle.

Ejection Fraction Level of Physical Conditioning

LV (Rest/Ex)

RV (Rest/Ex)

Peak Work Load (kpm/kg)

Sedentary Sedentary Active Sedentary Sedentary Sedentary Sedentary Active Active Sedentary Active Active

61/50 67/83 66/68 55/50 67/66 66/67 61/59 61/65 61/62 41/28 61/52 43/44

53/57 36/39 43/55 51/46 51/42 38/35 45/47 47/51 38/38 40/34 40/37 34/36

3.2 7.3 7.9 9.0 3.5 15.0 7.3 11.3 10.3 4.2 9.2 14.7

January i5, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51

TABLE III

Control • Subjects

Peak exercise

TABLE IV

Heart Rate Response (4- Standard Deviation, beats/min)

159 4- 17 ~

Group 1

i46 4- 271

Group 2

142 ~E 201

Systolic Blood Pressure Response ( ± Standard Deviation, mm Hg) Control Subjects

Group 1

Group 2

Group 3

Rest

122 ~: 15

119 :~ 16.5

1 1 5 7 i3

120 ~ 22"

Peak exercise

197 4- 32 t

158 4- 27 t

158 4- 25 t

146 4- 23t,$

Group 3

121 4- 9 t,t

295

" p <0 05 control subjects, Groups 1 and 2 versus Group 3 at rest, peak exercise; all groups. ~p <0•001 control subjects, Group 1 versus Group 3 at peak exercise.

* p = NS all groups at rest• t p <0.001 rest ~ Peak exercise; all groups• ~ p <0.001 control subjects versus Groups 1, 2, and 3 at peak exercise•

tilt, after in vivo labelling of red blood cells with 25 mCi of technetium-99m. Patients were imaged in the supine position using a portable gamma camera (Ohio Nuclear, Series Sigma 420) equipped with a general purpose parallel-hole collimator. The camera and an electrocardiographic gate (Ohio Nuclear) were interfaced to a dedicated computer (Digital Equipment Corpor ation 11/40 or 11/60). Rest studies were acquired for 200 cardiac cycles. Graded exercise studies were performed using a Quinton supine bicycle ergometer as previously described from our laboratory.4,5 Exercise was started at a work load of 200 kpm and increased in 200 to 300 kpm increments every 3 minutes to patient exhaustion. At each level of exercise, a stable heart rate was attained before a 2-minute acquisition period. Heart rate, blood pressure, and a single lead electrocardiogram (lead II) were monitored throughout the study. All control subjects and patients were encouraged to achieve peak exertion to the limit of symptoms in the supine posture. Data analysis: All radionuclide angiographic analysis was performed on an 11/40 or 11/60 computer (Digital Equipment Corporation) using Gamma II software. Data were initially viewed in an endless loop format to visually define the borders of both the right and the left ventricles throughout the cardiac cycle. The right and left ventricular end-diastolic and endsystolic frames were determined from time-activity curves generated from regions of interest surrounding both ventricles. Four regions of interest were assigned to both end-systolic and end-diastolic frames; the 2 ventricles and a left and right paraventricular background region as previously described.4. 5 Two experienced observers independently analyzed the studies, and the reproducibility of the data analysis was assessed by comparing their results. Interobserver correlation for both left and right ventricular ejection fractions in VSD patients as well as control subjects was excellent (r = 0.88).4,'~ Ejection fractions (EF) were calculated by the following equation: EF = EDCbc - ESCbc/EDCbc, where EDCbc = background corrected end-diastolic counts and ESCbc = background corrected end-systolic counts.

Statistical analysis of the change from rest to exercise in each group was carried out with the 2-tailed paired t test. Comparisons among the 4 groups were carried out with the 2-tailed unpaired Student's t test.

t p <0.001 rest ~

TABLE V

H e a r t rate and systolic blood pressure response ( T a b l e III): T h e resting heart rates (mean ± SD) were similar in Groups 1 and 2 compared with control subjects (81 ± 15 and 76 ± 15 versus 72 ± 14; p = not significant [NS]). However, Group 3 patients had a slightly higher resting heart rate (96 ± 5; p <0.05 comp a r e d with control subjects, Groups I and 2). All s t u d y groups d e m o n s t r a t e d an increase in heart rate with exercise. However, Group 3 patients had the lowest peak exercise heart rate (121 ± 9; p <0.001 c o m p a r e d with control subjects and G r o u p 1). Resting systolic blood pressure was similar in the 4 s t u d y groups (Table IV). P e a k exercise systolic blood pressure (systolic B P ± SD) was again similar in Groups 1, 2, and 3 but was significantly higher in the control subjects (158 ± 27; 158 ± 25; 146 ± 23 versus 197 4- 32; p <0.001). W o r k load: T h e mean maximal work load achieved by control subjects (14 kpm/kg) was much higher than t h a t in any of the groups with VSD. T h e mean maximal work loads in Groups 1 and 2 were moderately reduced at 9.3 and 8.6 (p <0.001 versus control subjects) and markedly depressed in Group 3 at 4.1 kpm/kg (p <0.001 compared with control subjects; p <0.05 compared with Groups 1 and 2). As may be expected, patients who were active were able to reach significantly higher peak work loads t h a n sedentary patients (p <0.01 in Group 1 and p <0.003 in Group 2). Ejection fractions ( T a b l e V): Normal subjects: The m e a n left ventricular ejection fraction at rest (mean ejection fraction ± SD) was 0.70 ± 0.07 and increased to 0.79 ± 0.05 at peak exercise (p <0.001). T h e mean

Mean Ejection Fractions (-I- Standard Deviation) Control Subjects

LV RV

Results

Group 1

Group 2

Group 3

Rest

Exercise

Rest

Exercise

Rest

Exercise

Rest

Exercise

70 4- 7 46 4- 6

79 4- 5* 57 4- 4*

70 4- 7 44 4- 8

67 4- 6 46 4- 7

59 4- 9 t 43 4- 6

58 4- 14 43 4- 8

54 4- 6 t 30 4- 7 t

49 4- 2 34 4- 9

. * p <0.001 change in ejection fraction rest --~ peak exercise• ~ p <0.001 ejection fraction rest versus control rest• LV = left ventricle; RV = rtght ventricle•

296

VENTRICULAR FUNCTION IN VENTRICULAR SEPTAL DEFECT

% EJECTION FRACTION

%

L .V.

I00

EJECTION FRACTION

R V.

I00

90

L.V.

R V.

9O

80 8O

70

7O

60 6O 50 50 40 40 30

30

20 2O I0 PE4K EXERC/SE

REST 0

PE4K E X E R ( "~

REST

I0 PEAK

i

--

R~ST 0

E~E~C,SE J

PE4K

R~,T

E~E~,SE n

% EJECTION kRACTION

RV

L.V.

I00

80

70

60

5O

i

j /

/

40

30

2O

'°I REST 0

•

b

PEAK EXERCzSE i

RES T

PEAK EXERCISE

i

right ventricular ejection fraction at rest was 0.46 ± 0.06 and increased to 0.57 ± 0.04 at peak exercise (p

<0.001). Group 1 (Fig. 1, top left): Both left and right ventricular ejection fractions were similar to those in the control subjects at rest; however, both ventricles demonstrated a flat response to exercise, with a mean left ventricular ejection fraction of 0.70 ± 0.07 at rest and 0.67 ± 0.06 at peak exercise, and a mean right ventricular ejection fraction of 0.44 ± 0.08 at rest and 0.46 ± 0.07 at peak exercise (p <01001 versus control subjects at peak exercise for both left and right ventricles). When physical conditioning was taken into account, this trend was still found to exist (Table I). The left ventricular ejection fraction at rest was 0.72 ± 0.05 for

f

FIGURE 1. Left (L.V.) and right ventricular (R.V.) ejection fraction changes from rest to peak exercise. (Mean :E standard error of the mean for each ventricle are on the sides of the individual responses.) Top left, Group 1 (nonsurgical patients); lop right, Group 2 (surgical patients); bottom left, Group 3 (Eisenmenger's complex patients).

sedentary and 0.70 + 0.08 for active individuals. On exercise, both decreased to 0.71 ± 0.02 for sedentary and 0.66 ± 0.07 for active patients (p = NS between sedentary and active patients). Physical conditioning also did not account for an abnormal right ventricular ejection fraction response to exercise. Group 2 (Fig. 1, top right): The resting left ventricular ejection fraction was lower than that for control subjects (0.59 ± 0.09 versus 0.70 ± 0.07; p <0,001), whereas the right ventricular ejection fraction was similar to that of control subjects (0.43 + 0.06 versus 0.;t6 ± 0.06; p = NS). There was again a flat response to exercise for both ventricles, with a left ventricular ejection fraction at peak exercise of 0.58 ± 0.14 and a right ventricular ejection fraction at peak exercise of

January 15, 1983 THE AMERICAN JOURNAL OF CARDIOLOGY Volume 51

0.43 ~ 0.08 (p <0.001 for both ventricles compared with the control subjects). Physical conditioning again did not account for the marked deviation from the normal exercise response (Table II). Group 3 (Fig. 1, bottom left): Patients with Eisenmenger's complex had lower resting values for both !eft and right ventricles compared with control subjects (left ventricular ejection fraction 0.54 ± 0.06 versus 0.70 ± 0.07 and right ventricular ejection fraction 0.30 ± 0.07 versus 0.46 ± 0.06; p <0.001). The ejection fraction response to exercise for both ventricles was similarly flat or decreased (left ventricular ejection fraction 0.49 ± 0.02 and right ventricular ejection fraction 0.34 ± 0.09 at peak exercise; p <0.001 compared with control subjects). The mean ejection fraction response for both ventricles in all groups is displayed in Figure 2.

% EJ ECTI.ON FRACTION

t0£

L,V.

Discussion

R.V.

9C

8c 7(3

.lJ

60 5O

40 30 20

CONTROL ....... NON.SURGiCAL ............... SURGICAL ------EISENMENGERS

10 R~ST

Group 1: Patients with a small VSD (less than 2 to 1 pulmonary:systemic blood flow ratio with normal pulmonary vascular resistance) are believed to have a good overall prognosis with a functional status comparable to that of the general population. 6,7 In fact, a significant proportion may undergo spontaneous closure, and so far there is no evidenCe to suggest that an increase in shunt size can occur in these patients. 6-1i Our results show that patients with small shunts had normal resting values for left and right ventricular ejection fraction, but they significantly failed to increase with exercise. The exercise response could not be predicted on the basis of history, physical examination, electrocardiogram, or chest roentgenogram, as all patients were in NYHA class I and had essentially normal electrocardiograms and chest X-rays. The mean maximal work 10ad attained was significantly lower than that of the control group; however, systolic blood pressure and heart rate response were normal in these patients. This finding could not be explained by a sedentary level of physical conditioning, which was present in only a Small number of patients (4 of 17). This decrease in ejection fraction also did not correlate with age of patients, size of shunt, physical conditioning, or peak work load. Several explanations for these findings are possible. The ventricles may respond to exercise by dilating and increasing their end-diastolic volumes during exercise with a constant ejection fraction. The stroke volume would increase during exercise as the end-diastolic volume increased; however, this increase could indeed represent previously unsuspected myocardial dysfunction secondary to chronic volume overload. G r o u p 2: All operated patients in our study had surgery through a right ventricular approach. All operations were performed before the advent of cardioplegia and modern myocardial protection, which may well be a significant factor influencing postoperative myocardial function. Furthermore, 10 of 12 patients had surgery after the age of 2 years, and the duration of Uncorrected hem0dynamic insult in patients with VSD may also be an important factor in predicting outCOme.12,13The level of physical conditioning again did

297

PEAK

PEAK EXERCISE

0

FIGURE 2. Mean I~ft (L.V.) and right ventricular (R.V.) ejection fraction changes from rest to peak exercise in all groups compared with the control subjects. All study groups demonstrated a flat exercise response for both ventricles, whereas the ejection fraction increased with exercise in both ventricles in the control subjects. Control group, unbroken line; Group 1, short dashes; Group 2, dotted line; Group 3, long dashes.

not appear to explain the abnormalities in exercise ejection fraction (Table V). Several investigators have previously commented on impaired ventricular function after surgical repair of a congenital VSD. Jarmakani et aP 4 looked at the velocity-pressure relation as a measure of left ventricular contractility in postoperative patients an average of 1.3 years after successful surgical closure. The ejection fraction was slightly depressed at rest in all patients studied. The same investigators looked at pre- and postoperative catheterization data in 23 patients to study the effect of VSD closure on left heart volume, mass, and contractility. 15 The postoperative catheterizations were performed 1 to 5 years after surgery, with a mean time lapse of 2 years. They found that although there had been significant regression in left ventricular volume and mass postoperatively, the results were still significantly higher than the normal values. Furthe~more, postoperative resting left ventricular ejection fraction was lower than both the preoperative left ventricular ejection fraction and that of their normal control subjects (0.54 versus 0.60 and 0.63, respectively). They speculated that partially irreversible changes may be present in the myocardium associated with the hemodynamic effects of VSD. Maron et aP 6 studied the cardiac output and pulmonary arterial pressure response to upright exercise in 11 asymptomatic postoperative patients. Exercise studies were carried out 3 to 15 years (mean 9.5) after operative closur e of isolated VSD. Ten of the :l1 patients had had routine postoperative catheterizations within the first years after surgery and none had had evidence of residual shunt. Postoperative mean pulmonary arterial pressure was normal or mildly elevated at rest in

298

VENTRICULAR FUNCTION iN VENTRICULAR SEPTAL DEFECT

10 patients. During upright exercise sufficient to lower pulmonary arterial oxygen saturation to 30%, cardiac output was below the normal range in 5 patients, each of whom had been operated on after the age of 10 years. The investigators suggested that late cardiovascular function may be abnormal after repair of VSD and that the abnormalities may be related to age at operation. Ahmad and Hallidie-Smith 17 found no abnormality of ejection fraction at rest in the early postoperative period in their study patients, and they speculated that abnormalities of postoperative myocardial function suggested by earlier studies may be related to the unavailability of currently used intraoperative myocardial preservation techniques. Further follow-up data on this group of patients will be necessary before this controversy can be settled. However, this finding is still clinically important, as a large number of these patients are being followed up by cardiologists. Therefore, baseline myocardial function determinations are important for future comparison if any change in clinical status ocCURS.

Group 3: At one extreme of the spectrum of population with chronic ventricular volume overload is the patient with Eisenmenger's complex. These patients had an extremely poor resting and exercise response as measured by all variables including heart rate and blood pressure response, maximal work load attained, and ejection fractions of both ventricles. These patients represent the end-stage of biventricular volume overload and pressure overload on the right ventricle as well as the added detrimental effect of right to left shunting of desaturated blood supplying hypertrophied viable myocardium. In summary, with the aid of nuclear angiography, these data suggest the following: (1) The response of blood pressure, work load, and ejection fraction of both left and right ventricles to exercise in all patients with residual persistent VSD is abnormal compared with that of control subjects. (2) Rest and exercise left Veritricular ejection fractions may be impaired postoperatively, and surgical closure does not appear to normalize ventricular function in patients operated on in the remote past or after the age of 2 years, or both. (3) Resting

left and right ventricular function are abnormal in pa. tients with Eisenmenger's complex. (4) Lifelong chronic volume overload, even to a small degree, may be detri. mental to myocardial function. References 1. Bruce RA, John GG. Effects of upright posture and exercise onpulmonary hemodynamics in patients with central cardiovascular shunts, uirculation 1957;16:776-783. 2. Wackers FJT, Berger HJ, Johnstone DE, Goldman L, Reduto LA, Langon RA, Gottschalk A, Zaret BL, Multiple gated cardiac blood pool imaging for left ventricular ejection fraction: validation of the technique and assessment of variability. Am J CardJol 1979;43:1159-1166, 3. Maddahi J, Berman DS, Matsuoka DT, Waxman AD, Siankus KE, Ferrester JS, Swan HJC. A new technique for assessing right ventricular ejection fraction using rapid multiple-gated equilibrium cardiac blood pool scintig. raphy. Circulation 1979;60:581-589. 4. Bar-Shlomo BZ, Druck MN, Morch JE, Jablonsky G, Hilton JD, Feiglin DHI, McLaughlin PR. Exercise left ventricular function in trained and untrained healthy subjects. Circulation 1982;65:484-488. 5. Benson LN, Bonet J, McLaughlin PR, Olley PM, Feignn DHI, Druck MN, Trusler G, Rowe R, Morch JE. Assessment of right ventricular function during supille bicycle exercise after Mustard's operation. Circulation 1982;65:1052-1059. 6. Bloomfield DK. The natural history of ventricular septal defect in patients surviving infancy. Circulation 1964;29:914-955. 7. Corone P, Doyen F, Gaudeau S, Guerin F, Vernant P, Duncan H, Rumeau-Rouquette C, Gaudeul P. Natural history of ventricular septal defect: a study involving 790 cases. Circulation 1977 55:908-915. 8. Campbell M. Natural history of ventricular septal defect. Br Heart J 1971;33:246-257. 9. Keilh JD, Rose V, Collins G, Kidd BSL. Ventricular septal defect: incidence, morbidity and mortality in various age groups. Br Heart J 1971;33:8183. 10. Walker WJ, Garcia-Gonzalez E, Hall RJ, Czarnecki SW, Franklin RB, Das SK, Cheiilin MD. Interventricular septal defect: analysis of.415 catheterized cases, ninety with serial hemodynamic studies. Circulation 1965 31:5465, 11. Arcilla RA, Agustsson MH, Bicoff JP, Lynfield J, Weinberg M Jr, Fell EH, Gasul BM. Further observations on the natural history of isolated ventricular septal defects in infancy and childhood: serial cardiac catheterization studies in 75 patients. Circulation 1963;28:560-571. 12. Hallidie-Smith KA, Wilson RSE, Hart A, Zeidifard E. Functional status of patients with large ventricOlar septal defect and pulmonary vascular disease 6 to 10 years after surgical closure of their defects in childhood. Br Heart J 1977;39:1093-1101. 13. Sigmann JM, Perry BL, Behrendt DM, Stern AM, Kirsh MM, Sloan HE. Ventricular septal defect: results after repair in infancy. Am J Cardiol 1977;39:66-71. 14. Jarmakani JM, Graham TP Jr, Conent RV Jr. Left ventricular contractile state in children with successfully corrected ventricular septal defect. Circulation 1972;65,66:Suppl 1:102-110. 15. Jarmakani JM, Graham TP Jr, Canent RV Jr, Capp MP. The effect of corrective surgery on left heart volume and mass in children with ventricular septal defect. Am J Cardio11971;27:254-263. 16. Maron BJ, Redwood DR, Hirchfeld JW Jr, Goldstein RE, Morrow AG, Epstein SE. Postoperativ e assessment of patients with ventricular septal defect and pulmonary hypertension: response to intense upright exercise. CircUlation 1973;48:864-874. 17. Ahmad M, Hallidie-Smith KA. Assessment of left-to-right shunt and left ventricular function in isolated ventricular septal defect. Br Heart J 1979; 41:147-158.