Pediatr Cardiol 11:61-68, 1990
Pediatric Cardiology 9 Springer-VerlagNew York Inc. 1990
Original Articles Left Ventricular Contractile State After Surgical Correction of Tetralogy of Fallot: Risk Factors for Late Left Ventricular Dysfunction Gerd Hausdorf, 1 Claus Hinrichs, I Christoph A. Nienaber, 2 Claudia Schark, ~ and Ernst W. Keck I Departments of ~Pediatric Cardiology and 2Cardiology, University Hospital Hamburg, Hamburg, FRG
SUMMARY. The purpose of this study was to analyze potential "risk-factors" for late left ventricular dysfunction after surgical correction of Fallot's tetralogy (FT). As the ejection-phase indices cannot distinguish abnormalities of contractility from altered loading conditions, the slope values of the end-systolic pressure-length and stress-shortening relationships were analyzed by increasing afterload. Thirty-two patients were studied after surgical correction of Wl" in infancy. The age at investigation was 19.2 _+ 5.6 years, total correction had been performed at the age of 7.7 -+ 3.3 years. In 20 patients a one-stage operation was performed, and in 12 patients a two-stage correction. The control group consisted of 30 healthy volunteers, aged 1830 years. The following potential risk factors for left ventricular dysfunction were evaluated: one-stage vs. two-stage correction, age at total correction, preoperative systemic oxygen saturation, preoperative hematocrit, occurrence of hypoxic spells, preoperative ratio of left-to-right ventricular peak systolic pressure, and preoperative ratio of left-to-right ventricular end-diastolic volume. In most patients the baseline data for end-systolic wall stress lay outside the normal range, indicating abnormal loading conditions. Thus, analysis of load-independent indices of the contractile state seems to be mandatory in these patients. Our data show that the severity of preoperative hypoxemia is an important risk factor for late dysfunction of the left ventricle (p < 0.01). Additionally, the relation of left and right ventricular peak systolic pressures and enddiastolic volumes were related to the contractile state (p < 0.01). No influence of preoperative hypoxic spells, the need for a palliative aortopulmonary shunt, or the age at surgical correction on the postoperative contractile state was demonstrated. The latter may have been due to the fact that none of the patients were operated on within the first 2 years of life. KEY WORDS: Cardiac surgery - - Contractile state - - Echocardiography - - Hypoxemia - Tetralogy of Fallot
Cardiac performance is of increasing importance for the adolescent and adult with surgically corrected congenital heart disease because life-expectancy will have been prolonged. Tetralogy of Fallot (FT) is characterized predominantly by abnormalities of the right heart [16]. However, left ventricular size is often reduced due to diminished pulmonary blood flow [20]. This "hypoplasia" of the left ventricle could be of major importance for the long-term prognosis and quality of life after surgical correction of FT [5, 15, 28]. Additionally, the degree of Address offprint requests to: Dr. Gerd Hausdorf, Deutsches Herzzentrum Berlin, Abteilung Angeborene Herzfehler, Augustenburger Platz 1, D-1000 Berlin 65, FRG.
preoperative hypoxemia, an increased preoperative hematocrit, and the age at surgical correction could be potential "risk factors" for late left ventricular dysfunction [5, 10, 28]. The purpose of this study was to analyze potential risk factors for late left ventricular dysfunction after surgical correction of FT. Assessment of left ventricular performance is still a subject of controversy. Usually, ejection-phase indices--ejection fraction, fractional shortening or mean fiber shortening velocity--are used for evaluating left ventricular systolic performance. As these indices cannot distinguish abnormalities of contractility from altered loading conditions, their application in patients with altered loading conditions could well be
62
inadequate [6, 4, 38]. The relationship between endsystolic pressure and dimension has gained increased acceptance as a load-independent parameter of the contractile state [3, 25, 26, 32, 33, 36]. However, the relationship between end-systolic wall stress and fractional shortening or mean fiber shortening velocity proved to be even more sensitive to changes of the contractile state than the endsystolic pressure-dimension (Pes-Des) relationship, reflecting the dependence of the ejection-phase indices on the loading conditions [4, 6].
Pediatric Cardiology Vol. 11, No. 2, 1990
Study Protocol All examinations were performed at rest in the semisupine position. The patients relaxed for at least 10 min prior to the examination. Atropine (0.01 mg/kg) was administered intravenously prior to the examination to reduce reflex cardiac slowing [3, 4, 6, 17]. After baseline recordings the blood pressure was slowly increased by the infusion of angiotensin II (0.5-3.5 p~g/min). Systolic blood pressure was increased for 30 mmHg above baseline values. During the elevation of blood pressure several simultaneous recordings of the M-mode echocardiogram, carotid pulse tracing, and arterial blood pressure were performed. Informed consent was obtained from all patients or their parents.
Patients and Methods
Patients Thirty-two patients were studied after surgical correction of FT in infancy. The age at investigation was 19.2 -+ 5.6 years, total correction having been performed at the age of 7.7 + 3.3 years. In 20 patients a one-stage operation was performed (group 1), in 12 patients a two-stage correction was performed (group 2).
Control Group The control group consisted of 30 healthy volunteers, aged 18-30 years [17]. None had a history of cardiovascular disease. Physical examination, electrocardiography, M-mode, and two-dimensional echocardiography revealed no abnormalities.
Measurements and Calculation For all measurements three consecutive cardiac cycles were evaluated and the mean taken for further calculations. Data points were excluded when the heart rate varied by more than 10 beats/min from baseline values [3, 4, 6, 17]. The end-diastolic internal LV dimension (Ded) was measured at the beginning of the QRS complex of the electrocardiogram, while the end-systolic LV dimension (De0 and posterior wall thickness (PW~0 were measured at end-ejection as assessed from the time-corrected carotid pulse tracings. Fractional shortening (FS) was calculated as: Ded -- Des FS = - - . Oed
100%
End-systolic meridional wall stress (o'es) was calculated according to Brodie et al. [8]:
Echocardiograms These were recorded with a Picker 80C Echoview ultrasound imaging device using a 3.5 MHz and 2.25 MHz transducer. The M-mode echocardiograms were recorded from the tip of the mitral valve leaflets in standard position with a paper speed of 100 mm/s [34]. The transducer was held in position for all measurements. All echocardiograms were recorded by one echocardiographer to eliminate interobserver variability. Simultaneously with the M-mode echocardiogram, a carotid pulse tracing was recorded, which was corrected for pulse transmission delay by aligning it to the aortic valve echocardiogram. To evaluate the geometry of the left ventricular short axis, cross-sectional echocardiograms of the left ventricular short axis were performed.
Measurement of End-Systolic Pressure The systolic and diastolic blood pressure was measured by a Dinamap 845 vital Signs Monitor (Criticon Inc.). This device has been shown to estimate central aortic pressure accurately over a wide range of pressures, independent of cardiac index, systemic vascular resistance, heart rate, and body surface area [2]. The end-systolic pressure ( P J was determined by the method of Stefadouros et al. [35], using linear interpolation of the height of the dicrotic notch of the indirect carotid pulse tracing [2, 35].
o'es = 1.35.
PW~s 9 Des (g/cm z) 4 9 PWes 9 (1 + PWes/D~)
The slopes of the relationships between end-systolic pressure and dimension (Pes-De~ relationship), and end-systolic stress and fractional shortening (o'es-FS relationship) were calculated a c cording to the following equations: Pes-Des relationship: o-e~-FS relationship:
Pes = al " Des + b FS = a 2- O-es+ FS0
Preoperative Parameters To analyze potential preoperative risk factors for late left ventricular performance, the following parameters were evaluated: 1. 2. 3. 4. 5. 6.
One-stage vs. two-stage correction. Age at surgical correction. Preoperative systemic oxygen saturation. Preoperative hematocrit. Occurrence of hypoxic spells. Left ventricular peak systolic pressure in percent of the right ventricular peak systolic pressure. 7. Left ventricular end-diastolic volume in percent of the right ventricular end-diastolic volume.
G. Hausdorf et al.: Contractile State After Correction of Fallot's Tetralogy
63
Table 1. Preoperative parameters: influence of one- and two-stage operation
n Age(years) Age at OP Time since OP (years) pLV/pRV (%) O2-Sat (%) Hct (%) LV/RVEDV (%)
All patients
Group 1
Group 2
30 19 7.7 11.8 102 83.5 46 88
12 22 9.4 12.0 99 83.6 47 92
18 18 6.9 11.8 104 83.6 46 87
-+ 5.4 _+ 3.2 -+ 3.6 -+ 13.7 -+ 9.7 -+ 10.7 +- 28
-+ 6.1 -+ 2.9 a -+ 4.2 -+ 7.7 -+ 5.9 -+ 6.6 +- 38
-+ 4.9 -+ 3.1 -+ 3.4 -+ 15.0 -+ 11.0 -+ 12.5 -+ 22
pLV/RV%, left ventricular peak systolic pressure in percent of right ventricular peak systolic pressure; OrSat, systemic oxygen saturation; Hct, hematocrit; LV/RVEDV, left ventricular end-diastolic volume in percent of the right ventricular end-diastolic volume; OP, operation. Only the age at surgical correction differed significantly between patients with one- and two-stage operation (p < 0.05). ~p < 0.05.
Assessment of Right and Left Ventricular Volumes For volume determinations no extra-beats or postextrasystolic beats were analyzed; because of this, the preoperative anglograms of five patients had to be excluded from the study. The end-diastolic conteur of the right and left ventricular angiograms were manually digitized using a Cardio 200 (Kontron Image Analysis) computer. The left ventricular end-diastolic volume was calculated by the area-length method [13], using the regression equation according to Graham et al. [11] for correction of the measured volumes. The right ventricular end-diastolic volume was calculated by Simpson's rule using the regression equation according to Graham et al. [14] for correction of the measured volumes. No estimation of absolute end-diastolic volumes was performed due to inadequate calibrations of some of the angiograms. As the loading conditions at the time of cardiac catheterization were unknown and can assumed to have altered [20], the right and left ventricular ejection fractions were not analyzed.
systemic oxygen saturations did not differ before total correction in both groups (Table 1). The preoperative patient data are given in Table 1. There was no difference of the age at investigation between both groups, while the age at operation was significantly lower in group 1 (p < 0.05). In two patients the echocardiograms were technically inadequate, so that these patients were excluded from further analysis. Flat septal movement was observed in 12 patients. In one patient septal movement was paradoxical: ejection-phase indices were not calculated in this patient. Cross-sectional echocardiography revealed a circular shape of the left ventricular short axis in all patients studied. In 14 patients hypoxic spells had been observed before surgical correction; in five of these a twostage correction was performed. The preoperative parameters for patients with and without hypoxic spells are given in Table 2.
Statistics Linear regression analysis was performed according to Pearson. When indicated, partial regression analysis was performed. Differences between group means were analyzed by an unpaired t test. A value o f p < 0.05 was regarded as statistically significant.
Results
In 20 patients (group 1) a one-stage operation had been performed, and in 12 patients a two-stage operation was performed (group 2). The aortopulmonary shunt resulted in a significant increase of systemic oxygen saturation from 63 -+ 18.5% to 83.6 -+ 11.0% in group 2 patients (p < 0.01). Thus,
Parameters of the Contractile State The individual data for end-systolic stress and fractional shortening at baseline conditions are shown in Fig. 1. Although the mean value for both parameters did not differ significantly from the mean value in the control group (Table 3), the standard deviation was much higher in the patient group. Thus, most of the data points are outside of the normal range (Fig. 1), indicating abnormal loading conditions. The mean values for the slopes of the end-systolic pressure-dimension relationship and relationship between end-systolic stress and fractional
64
Pediatric Cardiology Vol. 11, No. 2, 1990
Table 2. Preoperative parameters: influence of hypoxic spells on late systolic function
n Age(years) Age at OP (years) Time since OP (years) p L V / p R V (%) O2-Sat (%) Hct (%) L V / R V E D V (%)
All patients
H y p o x i c spells
No hypoxic spells
30 19 7.7 11.8 102 83.5 46 88
14 17 6.2 11.2 103 80.8 48 88
16 21 8.7 12.2 101 85.5 45 89
• 5.4 • 3.2 + 3.6 • 13.7 • 9.7 _+ 10.7 • 28
• 4.0 • 3.0 _+ 3.3 • 11.9 • 8.1 • 11.7 • 31
• • -+ • -+ • •
5.6 3.0 3.9 15.2 10.4 10.1 26
See Table 1 for abbreviations. N o significant differences due to preoperative hypoxic spells were observed.
Table 3. P a r a m e t e r s of the contractile state late after surgical correction of tetralogy of Fallot (FT)
FSbaseline(%) O'baseline(g cm -2) Pes-Des (cm m m H g -1) ~res-FS (cm 2 g ])
All patients
Group 1
31.1 51.8 46.2 -0.272
30.0 50.1 43.4 -0.267
-+ • _+ •
6.9 19.6 16.2 0.112
Group 2
+• • -+
5.6 11.8 13.5 0.090
31.6 52.9 47.0 -0.273
+ • -+ -+
Control group 7.8 22.9 17.5 0.127
34.1 46.8 54.1 -0.249
-+ -+ • •
1.9 5.9 18.4 0.060
FSbaseline, fractional shortening at baseline conditions; O'base~,e,end-systolic wall stress at baseline conditions; Pes-Des, slope value of the end-systolic p r e s s u r e - d i m e n s i o n relationship; o'e~-FS, slope value of the relationship b e t w e e n end-systolic stress and fractional shortening.
The slope values were slightly reduced in patients with preoperative hypoxic spells, but no significant differences of the contractile state indices could be shown with regard to the preoperative occurrence of hypoxic spells (Table 4).
FS
SO-
Regression Analysis of Preoperative "Risk Factors"
NORMAL RANGE
3O
el
The linear regression coefficients for the relationships of potential risk factors and the parameters of left ventricular performance are given in Table 5A, while table 5B gives the relationships between the potential risk factors.
N
NORMAL RANGE
SO
Oes
IOO g/cm
2
Fig. 1. T h e individual points for end-systolic wall stress and fractional shortening at baseline conditions (patient group) are illustrated together with the m e a n s and 95% confidence limits of these indices in the control group.
shortening are given in Table 3. Although the slope values were slightly reduced in the patient groups, no significant differences were observed in comparison with the control group.
Analysis of Partial Regressions Because of the significant correlations between some of the potential risk factors, partial regressions were calculated as shown in Table 6. Analysis of partial regressions showed the slope value of the end-systolic pressure-dimension relationship to be significantly dependent on the preoperative systemic oxygen saturation and the preoperative left ventricular peak systolic pressure as a percentage of the right ventricular peak systolic pressure (Table 6, Fig. 2A). The slope value of the relationship
G. Hausdorf et al.: Contractile State After Correction of Fallot's Tetralogy
65
Table 4. Indices of the contractile state for patients with and without preoperative bypoxic spells
Pc~-D~ (cm mmHg ~) o-~-FS (cm 2 g 1)
All patients
Hypoxic spells
No hypoxic spells
46.2 + 16.2 -0.272 -+ 0.112
44.1 _+ 14.5 -0.298 -+- 0.122
47.8 - 17.3 -0.256 -+ 0.101
See Table 3 for abbreviations.
Table 5, A. Correlations between potential "risk factors" and parameters of left ventricular performance
Pes-Des Age Age at OP Time pL/RV% O2-Sat Hct L/RVEDV
o'e~-FS
FSbase
O'base
0.4126 ~ 0.3736 ~ 0,3079 -0.1541 -0.3611 0.1171 0.2338
-0.2407 -0.2922 -0.1126 0.4063 ~ 0.3386 -0.0513 0.0348
0.2412 0.1365 0,2218 -0.3155 -0.3140 -0.0247 -0.1514
-0.1110 0.0624 -0.2492 =0.2863 0.5389 b -0.1541 0.2616
B. Correlations between the potential "risk factors"
Age at OP Time pL/RV% O:-Sat Hct L/RVEDV
Age
Age at OP
Time since OP
pL/RV%
O2-Sat
Hct
0.7474 ~ 0.790F 0.0909 -0.0929 -0.1336 -0.2040
0.195 0.2468 0.0340 -0.087 0.0041
0.2468 -0.1641 -0.1152 -0.2801
0.0187 -0.1391 0.0753
-0.5413 b 0.4599"
-0.0417
See Tables 1 and 3 for abbreviations. Time, time since OP. a p < 0.05; bp < 0.01; ~p < 0.001.
Table 6, Analysis of partial regressions
Predictor
Regressor
Excluded influence
Partial correlation coefficient (r)
Pcs-Des Pes-Des P~s-D,~ Pe~-Des o'es-FS o'e~-FS Cres-FS O'e~-FS (re~-FS o-~s-FS cr~s-FS o'e~-FS
O2-Sat pL/RV% L/RVEDV L/RVEDV Age at OP O2-Sat pL/RV% L/RVDEV O2-Sat pL/RV% L/RVEDV O2-Sat
pL/RV% O2-Sat O2-Sat pL/RV% Age Age Age Age pL/RV% O2-Sat O2-Sat L/RVEDV
0.568 b 0.559 b 0.018 0.075 0.108 -0.312 - 0.211 0.357 - 0.3682 -0.158 0.483 b -0.543 b
See table 5 for abbreviations. p < 0.05; b p < 0.01
between end-systolic stress and fractional shortening was significantly related to the preoperative systemic oxygen saturation and the preoperative left ventricular end-diastolic volume as a percentage of the end-diastolic right ventricular volume (Table 6, Fig. 2B).
Discussion After surgical correction of FT, the anatomy is reconstructed; however, cardiac problems are still evident. In a variable n u m b e r of patients, residual right ventricular outflow tract obstruction occurs
66
P e d i a t r i c C a r d i o l o g y V o l . 11, N o . 2, 1990
o , - FS
P~- D,,
o 6O
oo9 99
"
Fig. 2. A "Risk factors"for late left ventricular
,
dysfunction after surgical correction of tetralogy of Fallot (F73. The correlations between the preoperative oxygen-saturation and the slope value of the end-systolic pressure-length relationship (Pe~-Dc0 (partial regression: r = 0.568, p < 0.01), and the correlation between the preoperative ratio of the left and right ventricular peak pressure (pLV/RV%) (regression: r = 0.559, p < 0.01) are shown. Significant intercorrelations between the risk factors (Table 5B) afforded the calculation of partial regressions.
~-0_4
0
60 02-
8o
60
Saturation
0
2 --
80 Saturation
P~- D
o
9
0
60
0 60
100
p LV/RV~
-
50
FS
B Risk fuctor.~ for tute l~I/'t ventricuhu" ~Iv.~[~tnction ctjter surgical correction o[ FT. The correlations between the preoperative oxygen-saturation and the end-systolic stress-shortening (cr~,-FS) relationships (partial regression: r - 0.543, p < 0.01), and the correlation between the ratio of the left and right ventricular end-diastolic volume (LV/RVEDVC~) and the end-systolic stress-shortening relationship (partial regression: r = 0.483, p < 0.01: r - 0.559. p < 0.01) are shown. Significant correlations between the risk factors (Table 5B) made it necessary to calculate partial regressions.
100
LV/RVEDV~
[12, 21, 22, 31]. After right ventricular outflow tract reconstruction with a transannular patch, pulmonary regurgitation is regularly observed [22, 23, 31]. In a small number of patients, a residual ventricular septal defect is found [1, 9, 12, 31]. Rhythm disturbances and impaired left ventricular function seem to be of increasing importance for the long-term prognosis and quality of life in this condition [5, 10, 12, 28-30]. Previous studies of left ventricular performance after surgical correction of FT gave conflicting results. Jarmakani et al. [20] reported reduced left ventricular ejection fractions before and after surgical correction. While most workers [18, 19, 37] reported an impaired increase of cardiac output with exercise testing, others [7, 30] reported a normal response to exercise and normal resting ejection fractions. As the ejection-phase indices cannot distinguish abnormalities of contractility from altered loading conditions, their application to patients with altered loading conditions seems to be inadequate. In most of our patients, afterload was abnormal in terms of end-systolic wall stress (Fig. 1). For this reason the end-systolic pressure-dimension relationship was evaluated as a load-independent parameter of the contractile state [3, 25, 26, 32, 36] and, additionally, the relationship between end-systolic wall stress and fractional shortening. The latter proved to be even more sensitive to changes in the contractile
state than the end-systolic pressure-dimension relationship [4, 6]. However, the end-systolic stressshortening relationship is influenced by preload changes, although it incorporates afterload [17]. Our data strongly indicate that the severity of preoperative hypoxemia is an important risk factor for late left ventricular dysfunction (Fig. 2, Table 6). In contrast to De Lorgeril et al. [10], preoperative hypoxic spells were not shown to have a significant influence on the contractile state late after surgical correction of FT (Table 4). Also, the need for a palliative aortopulmonary shunt before total correction was not a significant risk factor. There were no differences in the preoperative hemodynamic parameters between patients with one-stage and twostage correction (Table 3), despite a younger age at total correction. These patients had been operated upon during a period when total correction was delayed, whenever possible. Group-2 patients had been judged to need earlier definitive correction, although hemodynamic data do not support this decision retrospectively. In addition to the degree of preoperative hypoxemia, the relation of left and right ventricular peak systolic pressures and end-diastolic volumes were related to the contractile state (Table 6, Fig. 2). Both parameters reflect preoperative pulmonary perfusion that is related to the development of the left ventricle in this condition [20, 24, 27]. It would seem from the studies of Borow et al.
G. Hausdorf et al.: Contractile State After Correction of Fallot's Tetralogy
3OZ
l r--i
L_q I-! S
10 y e a r s
Fig. 3. The relative frequencies of the age at operation are illustrated in percent. None of the patients had been operated on below the age of 2 years.
[5] and James et al. [19] that postoperative left ventricular performance is clearly related to age at operation when surgical correction is performed within the first 2 years of life. This could not be shown in the current study, probably because we did not include any patients with early definitive correction (within the first 2 years of life). Figure 3 shows the distribution of frequencies for the parameter "age at operation." There are two maxima, one at the age of 5 years, and another one at 10 years.
Conclusion Left ventricular systolic function late after surgical correction of FT is significantly dependent on the degree of preoperative hypoxemia and thus on preoperative pulmonary perfusion. Although the importance of early correction within the first 2 years of life was not studied, early definitive correction seems to improve left ventricular function significantly. This may be due to reduced hypoxic alteration of the left ventricular myocardium.
References 1. Arciniegas E, Farooki ZQ, Hakimi M, Perry BL, Green EW (1980) Early and late results of total correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 80:770-778 2. Borow KM, Newburger JW (1982) Non-invasive estimation of central aortic pressure using the oscillometric method for analyzing systemic artery pulsatile blood flow: comparative study of indirect systolic, diastolic, and mean brachial artery pressure with simultaneous direct ascending aortic pressure measurements. Am Heart J 103:879-886 3. Borow KM, Neumann A, Wynne J (1982) Sensitivity of endsystolic pressure-dimension and pressure-volume relations to the inotropic state in humans. Circulation 65:988-997
67
4. Borow KM, Arensman FW, Webb C, Radley-Smith R, Yacoub MH (1984) Assessment of left ventricular contractile state after anatomic correction of transposition of the great arteries. Circulation 69:106-112 5. Borow KM, Green LH, Castaneda AR, Keane JF (1980) Left ventricular function after repair of tetralogy of Fallot and its relationship to age at surgery. Circulation 61:11501158 6. Borow KM, Henderson C, Neumann A, Colan S, Grady S, Papish S, Goorin A (1983) Assessment of left ventricular contractility in patients receiving doxorubicin. Ann Intern Med 99:750-756 7. Bristow JD, Kloster FE, Lees MH, Menashe VD, Grisworld HE, Starr A (1970) Serial cardiac catheterizations and exercise hemodynamics after correction of tetralogy of Fallot. Circulation 41:1057-1066 8. Brodie BR, McLaurin LP, Grossman W (1976) Combined hemodynamic-ultrasonic method for studying left ventricular wall stress. A m J Cardiol 37:864-870 9. Chiariello L, Meyer J, Wukasch DC, Hallman GL, Cooley DA (1975) Intracardiac repair of tetralogy of Fallot. Fiveyear review of 403 patients. J Thorac Cardiovasc Surg 70:529-535 10. De Lorgeril M, Friedli B, Assimacopoulos A (1984) Factors affecting left ventricular function after correction of tetralogy of Fallot. Br Heart J 52:536-541 11. Dodge HT, Sandler H, Ballew DW, Lord JD (1960) Use of biplane angiocardiography for measurement of left ventricular volume in man. A m Heart J 60:762-773 12. Garson A, Nihill MR, McNamara DG, Cooley DA (1979) Status of the adult and adolescent after repair of tetralogy of Fallot. Circulation 59:1232-1240 13. Graham TP, Jarmakani JM, Canent RV, Morrow MN (1971) Left heart volume estimation in infancy and childhood. Circulation 43:895-904 14. Graham TP, Jarmakani JM, Canent RV, Morrow MN (1973) Right ventricular volume determinations in children. Circulation 47:144-153 15. Graham TP Jr, Faulkner S, Bender H Jr, Wender CM (1977) Hypoplasia of the left ventricle. Rare cause of postoperative mortality in tetralogy of Fallot. Am J Cardio140:454-457 16. Guntheroth WG, Kawabori I (1977) Tetrad of Fallot. In: Moss AJ, Adams FH, Emmanouilides GC (eds) Heart disease in infants, children, and adolescents. Williams & Wilkins, Baltimore, pp 276-288 17. Hausdorf G, Gluth J, Nienaber CA (1987) Non-invasive assessment of end-systolic pressure-length and stress-shortening relationships in normal individuals: significance of different loading conditions induced by methoxamine and angiotensin II. Eur Heart J 8:1099-1108 18. Hirschfeld S, Tuboku-Metzger AJ, Borkat G, Ankeney J, Clayman J, Liebman J (1978) Comparison of exercise and catheterization results following total surgical correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 75:446-451 19. James FW, Kaplan S, Schwartz DC, Chou TC, Sandker MJ, Naylor V (1976) Response to exercise in patients after total surgical correction of tetralogy of Fallot. Circulation 54:671679 20. Jarmakani JMM, Graham TP, Canent RV, Jewett PH (1972) Left ventricular function in children with tetralogy of Fallot before and after palliative or corrective surgery. Circulation 46:478-490 21. Kavey REW, Thomas FD, Byrum CJ, Blackman MA, Sondheimer HM, Bore E L (1984) Ventricular arrhythmias
68
22.
23.
24.
25.
26.
27.
28.
29.
Pediatric Cardiology Vol. 11, No. 2, 1990
and biventricular dysfunction after repair of tetralogy of Fallot. J A m Coil Cardiol 4:126-131 Kirklin JW, Blackstone EH, Pacifico AD, Kirklin JK, Bargeron LM Jr (1984) Risk factors for early and late failure after repair of tetralogy of Fallot and their neutralization. Thorac Cardiovasc Surg 32:208-214 Klinner W, Reichert B, Pfallar M, Hatz R (1984) Late results after correction of tetralogy of Fallot necessitating outflow tract reconstruction. Thorac Cardiovasc Surg 32:244-247 Lev M, Rimoldi HJA, Rowlatt UF (1964) The quantitative anatomy of cyanotic tetralogy of Fallot. Circulation 30:531538 Marsh JD, Green LH, Wynne J, Conne PF, Grossman W (1979) Left ventricular end-systolic pressure-dimension and stress-length relations in normal human subjects. Am J Cardiol 44:1311-1317 Mehmel HC, Stockins B, Ruffmann K, Ohlshausen K, Schuler G, Kuebler W (1981) The linearity of the end-systolic pressure-volume relationship in man and its sensitivity for the assessment of left ventricular function. Circulation 63:1216-1222 Miller GAH, Kirklin JW, Rahimtoola SH, Swan HJC (1965) Volume of the left ventricle in tetralogy of Fallot. Am J Cardiol 16:488-493 Nomoto S, Muraoka R, Yokota M, Aoshima M, Kyoku I, Nakano H (1984) Left ventricular volume as a predictor of postoperative hemodynamics and a criterion for total correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 88:389394 Rocchini AP, Keane JF, Freed MD, Castaneda AR, Nadas AS (1978) Left ventricular function following attempted surgical repair of tetralogy of Fallot. Circulation 57:778
30. Rosing DR, Borer JS, Kent KM, Maron B J, Seides SF, Morrow AG, Epstein SE (1978) Long-term hemodynamic and electrocardiographic assessment following operative repair of tetralogy of Fallot. Circulation 58(suppl 1):1-209 31. Ruzyllo W, Nihill MR, Mullins CE, McNamara DG (1974) Hemodynamic evaluation of 221 patients after intracardiac repair of tetralogy of Fallot. A m J Cardiol 34:565-576 32. Sagawa K (1981) Editorial: the end-systolic pressure-volume relation of the ventricle: definition, modification and clinical use. Circulation 63:1223-1227 33. Sagawa K, Suga H, Shoukas AA, Bakalar KM (1977) Endsystolic pressure/volume ratio: a new index of ventricular contractility. Am J Cardiol 40:748-753 34. Sahn DJ, DeMaria A, Kissio J, Weyman A (1978) The committee on M-mode standardization of the American Society of Echocardiography. Recommendations regarding quantification in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 58:10721083 35. Stefadouros MA, Dougherty MJ, Grossman W,. Craig E (1973) Determination of systemic vascular resistance by a noninvasive technique. Circulation 47:101-107 36. Suga H, Sagawa K, Kostiuk D (1976) Controls of ventricular contractility assessed by pressure-volume ratio Emax. Cardiovasc Res 10:582-592 37. Wessel HU, Cunningham WJ, Paul MH, Bastanier CK, Muster AJ, Idriss FS (1980) Exercise performance in tetralogy of Fallot after intracardiac repair. J Thorac Cardiovasc Surg 80:582-593 38. Weber KT, Janicki JS (1980) The dynamics of ventricular contraction: force, length, and shortening. Fed Proc 39:188195
Pediatric Cardiology 9 Springer-VerlagNew York Inc. 1990
Original Articles Left Ventricular Contractile State After Surgical Correction of Tetralogy of Fallot: Risk Factors for Late Left Ventricular Dysfunction Gerd Hausdorf, 1 Claus Hinrichs, I Christoph A. Nienaber, 2 Claudia Schark, ~ and Ernst W. Keck I Departments of ~Pediatric Cardiology and 2Cardiology, University Hospital Hamburg, Hamburg, FRG
SUMMARY. The purpose of this study was to analyze potential "risk-factors" for late left ventricular dysfunction after surgical correction of Fallot's tetralogy (FT). As the ejection-phase indices cannot distinguish abnormalities of contractility from altered loading conditions, the slope values of the end-systolic pressure-length and stress-shortening relationships were analyzed by increasing afterload. Thirty-two patients were studied after surgical correction of Wl" in infancy. The age at investigation was 19.2 _+ 5.6 years, total correction had been performed at the age of 7.7 -+ 3.3 years. In 20 patients a one-stage operation was performed, and in 12 patients a two-stage correction. The control group consisted of 30 healthy volunteers, aged 1830 years. The following potential risk factors for left ventricular dysfunction were evaluated: one-stage vs. two-stage correction, age at total correction, preoperative systemic oxygen saturation, preoperative hematocrit, occurrence of hypoxic spells, preoperative ratio of left-to-right ventricular peak systolic pressure, and preoperative ratio of left-to-right ventricular end-diastolic volume. In most patients the baseline data for end-systolic wall stress lay outside the normal range, indicating abnormal loading conditions. Thus, analysis of load-independent indices of the contractile state seems to be mandatory in these patients. Our data show that the severity of preoperative hypoxemia is an important risk factor for late dysfunction of the left ventricle (p < 0.01). Additionally, the relation of left and right ventricular peak systolic pressures and enddiastolic volumes were related to the contractile state (p < 0.01). No influence of preoperative hypoxic spells, the need for a palliative aortopulmonary shunt, or the age at surgical correction on the postoperative contractile state was demonstrated. The latter may have been due to the fact that none of the patients were operated on within the first 2 years of life. KEY WORDS: Cardiac surgery - - Contractile state - - Echocardiography - - Hypoxemia - Tetralogy of Fallot
Cardiac performance is of increasing importance for the adolescent and adult with surgically corrected congenital heart disease because life-expectancy will have been prolonged. Tetralogy of Fallot (FT) is characterized predominantly by abnormalities of the right heart [16]. However, left ventricular size is often reduced due to diminished pulmonary blood flow [20]. This "hypoplasia" of the left ventricle could be of major importance for the long-term prognosis and quality of life after surgical correction of FT [5, 15, 28]. Additionally, the degree of Address offprint requests to: Dr. Gerd Hausdorf, Deutsches Herzzentrum Berlin, Abteilung Angeborene Herzfehler, Augustenburger Platz 1, D-1000 Berlin 65, FRG.
preoperative hypoxemia, an increased preoperative hematocrit, and the age at surgical correction could be potential "risk factors" for late left ventricular dysfunction [5, 10, 28]. The purpose of this study was to analyze potential risk factors for late left ventricular dysfunction after surgical correction of FT. Assessment of left ventricular performance is still a subject of controversy. Usually, ejection-phase indices--ejection fraction, fractional shortening or mean fiber shortening velocity--are used for evaluating left ventricular systolic performance. As these indices cannot distinguish abnormalities of contractility from altered loading conditions, their application in patients with altered loading conditions could well be
62
inadequate [6, 4, 38]. The relationship between endsystolic pressure and dimension has gained increased acceptance as a load-independent parameter of the contractile state [3, 25, 26, 32, 33, 36]. However, the relationship between end-systolic wall stress and fractional shortening or mean fiber shortening velocity proved to be even more sensitive to changes of the contractile state than the endsystolic pressure-dimension (Pes-Des) relationship, reflecting the dependence of the ejection-phase indices on the loading conditions [4, 6].
Pediatric Cardiology Vol. 11, No. 2, 1990
Study Protocol All examinations were performed at rest in the semisupine position. The patients relaxed for at least 10 min prior to the examination. Atropine (0.01 mg/kg) was administered intravenously prior to the examination to reduce reflex cardiac slowing [3, 4, 6, 17]. After baseline recordings the blood pressure was slowly increased by the infusion of angiotensin II (0.5-3.5 p~g/min). Systolic blood pressure was increased for 30 mmHg above baseline values. During the elevation of blood pressure several simultaneous recordings of the M-mode echocardiogram, carotid pulse tracing, and arterial blood pressure were performed. Informed consent was obtained from all patients or their parents.
Patients and Methods
Patients Thirty-two patients were studied after surgical correction of FT in infancy. The age at investigation was 19.2 -+ 5.6 years, total correction having been performed at the age of 7.7 + 3.3 years. In 20 patients a one-stage operation was performed (group 1), in 12 patients a two-stage correction was performed (group 2).
Control Group The control group consisted of 30 healthy volunteers, aged 18-30 years [17]. None had a history of cardiovascular disease. Physical examination, electrocardiography, M-mode, and two-dimensional echocardiography revealed no abnormalities.
Measurements and Calculation For all measurements three consecutive cardiac cycles were evaluated and the mean taken for further calculations. Data points were excluded when the heart rate varied by more than 10 beats/min from baseline values [3, 4, 6, 17]. The end-diastolic internal LV dimension (Ded) was measured at the beginning of the QRS complex of the electrocardiogram, while the end-systolic LV dimension (De0 and posterior wall thickness (PW~0 were measured at end-ejection as assessed from the time-corrected carotid pulse tracings. Fractional shortening (FS) was calculated as: Ded -- Des FS = - - . Oed
100%
End-systolic meridional wall stress (o'es) was calculated according to Brodie et al. [8]:
Echocardiograms These were recorded with a Picker 80C Echoview ultrasound imaging device using a 3.5 MHz and 2.25 MHz transducer. The M-mode echocardiograms were recorded from the tip of the mitral valve leaflets in standard position with a paper speed of 100 mm/s [34]. The transducer was held in position for all measurements. All echocardiograms were recorded by one echocardiographer to eliminate interobserver variability. Simultaneously with the M-mode echocardiogram, a carotid pulse tracing was recorded, which was corrected for pulse transmission delay by aligning it to the aortic valve echocardiogram. To evaluate the geometry of the left ventricular short axis, cross-sectional echocardiograms of the left ventricular short axis were performed.
Measurement of End-Systolic Pressure The systolic and diastolic blood pressure was measured by a Dinamap 845 vital Signs Monitor (Criticon Inc.). This device has been shown to estimate central aortic pressure accurately over a wide range of pressures, independent of cardiac index, systemic vascular resistance, heart rate, and body surface area [2]. The end-systolic pressure ( P J was determined by the method of Stefadouros et al. [35], using linear interpolation of the height of the dicrotic notch of the indirect carotid pulse tracing [2, 35].
o'es = 1.35.
PW~s 9 Des (g/cm z) 4 9 PWes 9 (1 + PWes/D~)
The slopes of the relationships between end-systolic pressure and dimension (Pes-De~ relationship), and end-systolic stress and fractional shortening (o'es-FS relationship) were calculated a c cording to the following equations: Pes-Des relationship: o-e~-FS relationship:
Pes = al " Des + b FS = a 2- O-es+ FS0
Preoperative Parameters To analyze potential preoperative risk factors for late left ventricular performance, the following parameters were evaluated: 1. 2. 3. 4. 5. 6.
One-stage vs. two-stage correction. Age at surgical correction. Preoperative systemic oxygen saturation. Preoperative hematocrit. Occurrence of hypoxic spells. Left ventricular peak systolic pressure in percent of the right ventricular peak systolic pressure. 7. Left ventricular end-diastolic volume in percent of the right ventricular end-diastolic volume.
G. Hausdorf et al.: Contractile State After Correction of Fallot's Tetralogy
63
Table 1. Preoperative parameters: influence of one- and two-stage operation
n Age(years) Age at OP Time since OP (years) pLV/pRV (%) O2-Sat (%) Hct (%) LV/RVEDV (%)
All patients
Group 1
Group 2
30 19 7.7 11.8 102 83.5 46 88
12 22 9.4 12.0 99 83.6 47 92
18 18 6.9 11.8 104 83.6 46 87
-+ 5.4 _+ 3.2 -+ 3.6 -+ 13.7 -+ 9.7 -+ 10.7 +- 28
-+ 6.1 -+ 2.9 a -+ 4.2 -+ 7.7 -+ 5.9 -+ 6.6 +- 38
-+ 4.9 -+ 3.1 -+ 3.4 -+ 15.0 -+ 11.0 -+ 12.5 -+ 22
pLV/RV%, left ventricular peak systolic pressure in percent of right ventricular peak systolic pressure; OrSat, systemic oxygen saturation; Hct, hematocrit; LV/RVEDV, left ventricular end-diastolic volume in percent of the right ventricular end-diastolic volume; OP, operation. Only the age at surgical correction differed significantly between patients with one- and two-stage operation (p < 0.05). ~p < 0.05.
Assessment of Right and Left Ventricular Volumes For volume determinations no extra-beats or postextrasystolic beats were analyzed; because of this, the preoperative anglograms of five patients had to be excluded from the study. The end-diastolic conteur of the right and left ventricular angiograms were manually digitized using a Cardio 200 (Kontron Image Analysis) computer. The left ventricular end-diastolic volume was calculated by the area-length method [13], using the regression equation according to Graham et al. [11] for correction of the measured volumes. The right ventricular end-diastolic volume was calculated by Simpson's rule using the regression equation according to Graham et al. [14] for correction of the measured volumes. No estimation of absolute end-diastolic volumes was performed due to inadequate calibrations of some of the angiograms. As the loading conditions at the time of cardiac catheterization were unknown and can assumed to have altered [20], the right and left ventricular ejection fractions were not analyzed.
systemic oxygen saturations did not differ before total correction in both groups (Table 1). The preoperative patient data are given in Table 1. There was no difference of the age at investigation between both groups, while the age at operation was significantly lower in group 1 (p < 0.05). In two patients the echocardiograms were technically inadequate, so that these patients were excluded from further analysis. Flat septal movement was observed in 12 patients. In one patient septal movement was paradoxical: ejection-phase indices were not calculated in this patient. Cross-sectional echocardiography revealed a circular shape of the left ventricular short axis in all patients studied. In 14 patients hypoxic spells had been observed before surgical correction; in five of these a twostage correction was performed. The preoperative parameters for patients with and without hypoxic spells are given in Table 2.
Statistics Linear regression analysis was performed according to Pearson. When indicated, partial regression analysis was performed. Differences between group means were analyzed by an unpaired t test. A value o f p < 0.05 was regarded as statistically significant.
Results
In 20 patients (group 1) a one-stage operation had been performed, and in 12 patients a two-stage operation was performed (group 2). The aortopulmonary shunt resulted in a significant increase of systemic oxygen saturation from 63 -+ 18.5% to 83.6 -+ 11.0% in group 2 patients (p < 0.01). Thus,
Parameters of the Contractile State The individual data for end-systolic stress and fractional shortening at baseline conditions are shown in Fig. 1. Although the mean value for both parameters did not differ significantly from the mean value in the control group (Table 3), the standard deviation was much higher in the patient group. Thus, most of the data points are outside of the normal range (Fig. 1), indicating abnormal loading conditions. The mean values for the slopes of the end-systolic pressure-dimension relationship and relationship between end-systolic stress and fractional
64
Pediatric Cardiology Vol. 11, No. 2, 1990
Table 2. Preoperative parameters: influence of hypoxic spells on late systolic function
n Age(years) Age at OP (years) Time since OP (years) p L V / p R V (%) O2-Sat (%) Hct (%) L V / R V E D V (%)
All patients
H y p o x i c spells
No hypoxic spells
30 19 7.7 11.8 102 83.5 46 88
14 17 6.2 11.2 103 80.8 48 88
16 21 8.7 12.2 101 85.5 45 89
• 5.4 • 3.2 + 3.6 • 13.7 • 9.7 _+ 10.7 • 28
• 4.0 • 3.0 _+ 3.3 • 11.9 • 8.1 • 11.7 • 31
• • -+ • -+ • •
5.6 3.0 3.9 15.2 10.4 10.1 26
See Table 1 for abbreviations. N o significant differences due to preoperative hypoxic spells were observed.
Table 3. P a r a m e t e r s of the contractile state late after surgical correction of tetralogy of Fallot (FT)
FSbaseline(%) O'baseline(g cm -2) Pes-Des (cm m m H g -1) ~res-FS (cm 2 g ])
All patients
Group 1
31.1 51.8 46.2 -0.272
30.0 50.1 43.4 -0.267
-+ • _+ •
6.9 19.6 16.2 0.112
Group 2
+• • -+
5.6 11.8 13.5 0.090
31.6 52.9 47.0 -0.273
+ • -+ -+
Control group 7.8 22.9 17.5 0.127
34.1 46.8 54.1 -0.249
-+ -+ • •
1.9 5.9 18.4 0.060
FSbaseline, fractional shortening at baseline conditions; O'base~,e,end-systolic wall stress at baseline conditions; Pes-Des, slope value of the end-systolic p r e s s u r e - d i m e n s i o n relationship; o'e~-FS, slope value of the relationship b e t w e e n end-systolic stress and fractional shortening.
The slope values were slightly reduced in patients with preoperative hypoxic spells, but no significant differences of the contractile state indices could be shown with regard to the preoperative occurrence of hypoxic spells (Table 4).
FS
SO-
Regression Analysis of Preoperative "Risk Factors"
NORMAL RANGE
3O
el
The linear regression coefficients for the relationships of potential risk factors and the parameters of left ventricular performance are given in Table 5A, while table 5B gives the relationships between the potential risk factors.
N
NORMAL RANGE
SO
Oes
IOO g/cm
2
Fig. 1. T h e individual points for end-systolic wall stress and fractional shortening at baseline conditions (patient group) are illustrated together with the m e a n s and 95% confidence limits of these indices in the control group.
shortening are given in Table 3. Although the slope values were slightly reduced in the patient groups, no significant differences were observed in comparison with the control group.
Analysis of Partial Regressions Because of the significant correlations between some of the potential risk factors, partial regressions were calculated as shown in Table 6. Analysis of partial regressions showed the slope value of the end-systolic pressure-dimension relationship to be significantly dependent on the preoperative systemic oxygen saturation and the preoperative left ventricular peak systolic pressure as a percentage of the right ventricular peak systolic pressure (Table 6, Fig. 2A). The slope value of the relationship
G. Hausdorf et al.: Contractile State After Correction of Fallot's Tetralogy
65
Table 4. Indices of the contractile state for patients with and without preoperative bypoxic spells
Pc~-D~ (cm mmHg ~) o-~-FS (cm 2 g 1)
All patients
Hypoxic spells
No hypoxic spells
46.2 + 16.2 -0.272 -+ 0.112
44.1 _+ 14.5 -0.298 -+- 0.122
47.8 - 17.3 -0.256 -+ 0.101
See Table 3 for abbreviations.
Table 5, A. Correlations between potential "risk factors" and parameters of left ventricular performance
Pes-Des Age Age at OP Time pL/RV% O2-Sat Hct L/RVEDV
o'e~-FS
FSbase
O'base
0.4126 ~ 0.3736 ~ 0,3079 -0.1541 -0.3611 0.1171 0.2338
-0.2407 -0.2922 -0.1126 0.4063 ~ 0.3386 -0.0513 0.0348
0.2412 0.1365 0,2218 -0.3155 -0.3140 -0.0247 -0.1514
-0.1110 0.0624 -0.2492 =0.2863 0.5389 b -0.1541 0.2616
B. Correlations between the potential "risk factors"
Age at OP Time pL/RV% O:-Sat Hct L/RVEDV
Age
Age at OP
Time since OP
pL/RV%
O2-Sat
Hct
0.7474 ~ 0.790F 0.0909 -0.0929 -0.1336 -0.2040
0.195 0.2468 0.0340 -0.087 0.0041
0.2468 -0.1641 -0.1152 -0.2801
0.0187 -0.1391 0.0753
-0.5413 b 0.4599"
-0.0417
See Tables 1 and 3 for abbreviations. Time, time since OP. a p < 0.05; bp < 0.01; ~p < 0.001.
Table 6, Analysis of partial regressions
Predictor
Regressor
Excluded influence
Partial correlation coefficient (r)
Pcs-Des Pes-Des P~s-D,~ Pe~-Des o'es-FS o'e~-FS Cres-FS O'e~-FS (re~-FS o-~s-FS cr~s-FS o'e~-FS
O2-Sat pL/RV% L/RVEDV L/RVEDV Age at OP O2-Sat pL/RV% L/RVDEV O2-Sat pL/RV% L/RVEDV O2-Sat
pL/RV% O2-Sat O2-Sat pL/RV% Age Age Age Age pL/RV% O2-Sat O2-Sat L/RVEDV
0.568 b 0.559 b 0.018 0.075 0.108 -0.312 - 0.211 0.357 - 0.3682 -0.158 0.483 b -0.543 b
See table 5 for abbreviations. p < 0.05; b p < 0.01
between end-systolic stress and fractional shortening was significantly related to the preoperative systemic oxygen saturation and the preoperative left ventricular end-diastolic volume as a percentage of the end-diastolic right ventricular volume (Table 6, Fig. 2B).
Discussion After surgical correction of FT, the anatomy is reconstructed; however, cardiac problems are still evident. In a variable n u m b e r of patients, residual right ventricular outflow tract obstruction occurs
66
P e d i a t r i c C a r d i o l o g y V o l . 11, N o . 2, 1990
o , - FS
P~- D,,
o 6O
oo9 99
"
Fig. 2. A "Risk factors"for late left ventricular
,
dysfunction after surgical correction of tetralogy of Fallot (F73. The correlations between the preoperative oxygen-saturation and the slope value of the end-systolic pressure-length relationship (Pe~-Dc0 (partial regression: r = 0.568, p < 0.01), and the correlation between the preoperative ratio of the left and right ventricular peak pressure (pLV/RV%) (regression: r = 0.559, p < 0.01) are shown. Significant intercorrelations between the risk factors (Table 5B) afforded the calculation of partial regressions.
~-0_4
0
60 02-
8o
60
Saturation
0
2 --
80 Saturation
P~- D
o
9
0
60
0 60
100
p LV/RV~
-
50
FS
B Risk fuctor.~ for tute l~I/'t ventricuhu" ~Iv.~[~tnction ctjter surgical correction o[ FT. The correlations between the preoperative oxygen-saturation and the end-systolic stress-shortening (cr~,-FS) relationships (partial regression: r - 0.543, p < 0.01), and the correlation between the ratio of the left and right ventricular end-diastolic volume (LV/RVEDVC~) and the end-systolic stress-shortening relationship (partial regression: r = 0.483, p < 0.01: r - 0.559. p < 0.01) are shown. Significant correlations between the risk factors (Table 5B) made it necessary to calculate partial regressions.
100
LV/RVEDV~
[12, 21, 22, 31]. After right ventricular outflow tract reconstruction with a transannular patch, pulmonary regurgitation is regularly observed [22, 23, 31]. In a small number of patients, a residual ventricular septal defect is found [1, 9, 12, 31]. Rhythm disturbances and impaired left ventricular function seem to be of increasing importance for the long-term prognosis and quality of life in this condition [5, 10, 12, 28-30]. Previous studies of left ventricular performance after surgical correction of FT gave conflicting results. Jarmakani et al. [20] reported reduced left ventricular ejection fractions before and after surgical correction. While most workers [18, 19, 37] reported an impaired increase of cardiac output with exercise testing, others [7, 30] reported a normal response to exercise and normal resting ejection fractions. As the ejection-phase indices cannot distinguish abnormalities of contractility from altered loading conditions, their application to patients with altered loading conditions seems to be inadequate. In most of our patients, afterload was abnormal in terms of end-systolic wall stress (Fig. 1). For this reason the end-systolic pressure-dimension relationship was evaluated as a load-independent parameter of the contractile state [3, 25, 26, 32, 36] and, additionally, the relationship between end-systolic wall stress and fractional shortening. The latter proved to be even more sensitive to changes in the contractile
state than the end-systolic pressure-dimension relationship [4, 6]. However, the end-systolic stressshortening relationship is influenced by preload changes, although it incorporates afterload [17]. Our data strongly indicate that the severity of preoperative hypoxemia is an important risk factor for late left ventricular dysfunction (Fig. 2, Table 6). In contrast to De Lorgeril et al. [10], preoperative hypoxic spells were not shown to have a significant influence on the contractile state late after surgical correction of FT (Table 4). Also, the need for a palliative aortopulmonary shunt before total correction was not a significant risk factor. There were no differences in the preoperative hemodynamic parameters between patients with one-stage and twostage correction (Table 3), despite a younger age at total correction. These patients had been operated upon during a period when total correction was delayed, whenever possible. Group-2 patients had been judged to need earlier definitive correction, although hemodynamic data do not support this decision retrospectively. In addition to the degree of preoperative hypoxemia, the relation of left and right ventricular peak systolic pressures and end-diastolic volumes were related to the contractile state (Table 6, Fig. 2). Both parameters reflect preoperative pulmonary perfusion that is related to the development of the left ventricle in this condition [20, 24, 27]. It would seem from the studies of Borow et al.
G. Hausdorf et al.: Contractile State After Correction of Fallot's Tetralogy
3OZ
l r--i
L_q I-! S
10 y e a r s
Fig. 3. The relative frequencies of the age at operation are illustrated in percent. None of the patients had been operated on below the age of 2 years.
[5] and James et al. [19] that postoperative left ventricular performance is clearly related to age at operation when surgical correction is performed within the first 2 years of life. This could not be shown in the current study, probably because we did not include any patients with early definitive correction (within the first 2 years of life). Figure 3 shows the distribution of frequencies for the parameter "age at operation." There are two maxima, one at the age of 5 years, and another one at 10 years.
Conclusion Left ventricular systolic function late after surgical correction of FT is significantly dependent on the degree of preoperative hypoxemia and thus on preoperative pulmonary perfusion. Although the importance of early correction within the first 2 years of life was not studied, early definitive correction seems to improve left ventricular function significantly. This may be due to reduced hypoxic alteration of the left ventricular myocardium.
References 1. Arciniegas E, Farooki ZQ, Hakimi M, Perry BL, Green EW (1980) Early and late results of total correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 80:770-778 2. Borow KM, Newburger JW (1982) Non-invasive estimation of central aortic pressure using the oscillometric method for analyzing systemic artery pulsatile blood flow: comparative study of indirect systolic, diastolic, and mean brachial artery pressure with simultaneous direct ascending aortic pressure measurements. Am Heart J 103:879-886 3. Borow KM, Neumann A, Wynne J (1982) Sensitivity of endsystolic pressure-dimension and pressure-volume relations to the inotropic state in humans. Circulation 65:988-997
67
4. Borow KM, Arensman FW, Webb C, Radley-Smith R, Yacoub MH (1984) Assessment of left ventricular contractile state after anatomic correction of transposition of the great arteries. Circulation 69:106-112 5. Borow KM, Green LH, Castaneda AR, Keane JF (1980) Left ventricular function after repair of tetralogy of Fallot and its relationship to age at surgery. Circulation 61:11501158 6. Borow KM, Henderson C, Neumann A, Colan S, Grady S, Papish S, Goorin A (1983) Assessment of left ventricular contractility in patients receiving doxorubicin. Ann Intern Med 99:750-756 7. Bristow JD, Kloster FE, Lees MH, Menashe VD, Grisworld HE, Starr A (1970) Serial cardiac catheterizations and exercise hemodynamics after correction of tetralogy of Fallot. Circulation 41:1057-1066 8. Brodie BR, McLaurin LP, Grossman W (1976) Combined hemodynamic-ultrasonic method for studying left ventricular wall stress. A m J Cardiol 37:864-870 9. Chiariello L, Meyer J, Wukasch DC, Hallman GL, Cooley DA (1975) Intracardiac repair of tetralogy of Fallot. Fiveyear review of 403 patients. J Thorac Cardiovasc Surg 70:529-535 10. De Lorgeril M, Friedli B, Assimacopoulos A (1984) Factors affecting left ventricular function after correction of tetralogy of Fallot. Br Heart J 52:536-541 11. Dodge HT, Sandler H, Ballew DW, Lord JD (1960) Use of biplane angiocardiography for measurement of left ventricular volume in man. A m Heart J 60:762-773 12. Garson A, Nihill MR, McNamara DG, Cooley DA (1979) Status of the adult and adolescent after repair of tetralogy of Fallot. Circulation 59:1232-1240 13. Graham TP, Jarmakani JM, Canent RV, Morrow MN (1971) Left heart volume estimation in infancy and childhood. Circulation 43:895-904 14. Graham TP, Jarmakani JM, Canent RV, Morrow MN (1973) Right ventricular volume determinations in children. Circulation 47:144-153 15. Graham TP Jr, Faulkner S, Bender H Jr, Wender CM (1977) Hypoplasia of the left ventricle. Rare cause of postoperative mortality in tetralogy of Fallot. Am J Cardio140:454-457 16. Guntheroth WG, Kawabori I (1977) Tetrad of Fallot. In: Moss AJ, Adams FH, Emmanouilides GC (eds) Heart disease in infants, children, and adolescents. Williams & Wilkins, Baltimore, pp 276-288 17. Hausdorf G, Gluth J, Nienaber CA (1987) Non-invasive assessment of end-systolic pressure-length and stress-shortening relationships in normal individuals: significance of different loading conditions induced by methoxamine and angiotensin II. Eur Heart J 8:1099-1108 18. Hirschfeld S, Tuboku-Metzger AJ, Borkat G, Ankeney J, Clayman J, Liebman J (1978) Comparison of exercise and catheterization results following total surgical correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 75:446-451 19. James FW, Kaplan S, Schwartz DC, Chou TC, Sandker MJ, Naylor V (1976) Response to exercise in patients after total surgical correction of tetralogy of Fallot. Circulation 54:671679 20. Jarmakani JMM, Graham TP, Canent RV, Jewett PH (1972) Left ventricular function in children with tetralogy of Fallot before and after palliative or corrective surgery. Circulation 46:478-490 21. Kavey REW, Thomas FD, Byrum CJ, Blackman MA, Sondheimer HM, Bore E L (1984) Ventricular arrhythmias
68
22.
23.
24.
25.
26.
27.
28.
29.
Pediatric Cardiology Vol. 11, No. 2, 1990
and biventricular dysfunction after repair of tetralogy of Fallot. J A m Coil Cardiol 4:126-131 Kirklin JW, Blackstone EH, Pacifico AD, Kirklin JK, Bargeron LM Jr (1984) Risk factors for early and late failure after repair of tetralogy of Fallot and their neutralization. Thorac Cardiovasc Surg 32:208-214 Klinner W, Reichert B, Pfallar M, Hatz R (1984) Late results after correction of tetralogy of Fallot necessitating outflow tract reconstruction. Thorac Cardiovasc Surg 32:244-247 Lev M, Rimoldi HJA, Rowlatt UF (1964) The quantitative anatomy of cyanotic tetralogy of Fallot. Circulation 30:531538 Marsh JD, Green LH, Wynne J, Conne PF, Grossman W (1979) Left ventricular end-systolic pressure-dimension and stress-length relations in normal human subjects. Am J Cardiol 44:1311-1317 Mehmel HC, Stockins B, Ruffmann K, Ohlshausen K, Schuler G, Kuebler W (1981) The linearity of the end-systolic pressure-volume relationship in man and its sensitivity for the assessment of left ventricular function. Circulation 63:1216-1222 Miller GAH, Kirklin JW, Rahimtoola SH, Swan HJC (1965) Volume of the left ventricle in tetralogy of Fallot. Am J Cardiol 16:488-493 Nomoto S, Muraoka R, Yokota M, Aoshima M, Kyoku I, Nakano H (1984) Left ventricular volume as a predictor of postoperative hemodynamics and a criterion for total correction of tetralogy of Fallot. J Thorac Cardiovasc Surg 88:389394 Rocchini AP, Keane JF, Freed MD, Castaneda AR, Nadas AS (1978) Left ventricular function following attempted surgical repair of tetralogy of Fallot. Circulation 57:778
30. Rosing DR, Borer JS, Kent KM, Maron B J, Seides SF, Morrow AG, Epstein SE (1978) Long-term hemodynamic and electrocardiographic assessment following operative repair of tetralogy of Fallot. Circulation 58(suppl 1):1-209 31. Ruzyllo W, Nihill MR, Mullins CE, McNamara DG (1974) Hemodynamic evaluation of 221 patients after intracardiac repair of tetralogy of Fallot. A m J Cardiol 34:565-576 32. Sagawa K (1981) Editorial: the end-systolic pressure-volume relation of the ventricle: definition, modification and clinical use. Circulation 63:1223-1227 33. Sagawa K, Suga H, Shoukas AA, Bakalar KM (1977) Endsystolic pressure/volume ratio: a new index of ventricular contractility. Am J Cardiol 40:748-753 34. Sahn DJ, DeMaria A, Kissio J, Weyman A (1978) The committee on M-mode standardization of the American Society of Echocardiography. Recommendations regarding quantification in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 58:10721083 35. Stefadouros MA, Dougherty MJ, Grossman W,. Craig E (1973) Determination of systemic vascular resistance by a noninvasive technique. Circulation 47:101-107 36. Suga H, Sagawa K, Kostiuk D (1976) Controls of ventricular contractility assessed by pressure-volume ratio Emax. Cardiovasc Res 10:582-592 37. Wessel HU, Cunningham WJ, Paul MH, Bastanier CK, Muster AJ, Idriss FS (1980) Exercise performance in tetralogy of Fallot after intracardiac repair. J Thorac Cardiovasc Surg 80:582-593 38. Weber KT, Janicki JS (1980) The dynamics of ventricular contraction: force, length, and shortening. Fed Proc 39:188195
